/userhome/miniconda3/envs/mae/lib/python3.8/site-packages/torch/distributed/launch.py:178: FutureWarning: The module torch.distributed.launch is deprecated
and will be removed in future. Use torchrun.
Note that --use_env is set by default in torchrun.
If your script expects `--local_rank` argument to be set, please
change it to read from `os.environ['LOCAL_RANK']` instead. See 
https://pytorch.org/docs/stable/distributed.html#launch-utility for 
further instructions

  warnings.warn(
WARNING:torch.distributed.run:
*****************************************
Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. 
*****************************************
| distributed init (rank 1): env://, gpu 1
| distributed init (rank 5): env://, gpu 5
| distributed init (rank 7): env://, gpu 7
| distributed init (rank 0): env://, gpu 0
| distributed init (rank 3): env://, gpu 3
| distributed init (rank 4): env://, gpu 4
| distributed init (rank 2): env://, gpu 2
| distributed init (rank 6): env://, gpu 6
[10:57:51.880058] job dir: /userhome/mae-tmp
[10:57:51.880267] Namespace(aa='rand-m9-mstd0.5-inc1',
accum_iter=2,
batch_size=32,
blr=0.0006,
clip_grad=None,
color_jitter=None,
cutmix=0,
cutmix_minmax=None,
data_path='/dataset/ImageNet2012',
device='cuda',
dist_backend='nccl',
dist_eval=False,
dist_on_itp=False,
dist_url='env://',
distributed=True,
drop_path=0.1,
epochs=200,
eval=False,
finetune='',
global_pool=True,
gpu=0,
input_size=224,
layer_decay=1.0,
local_rank=0,
log_dir='./output_dir_cml_spikformer',
lr=None,
min_lr=1e-06,
mixup=0,
mixup_mode='batch',
mixup_prob=1.0,
mixup_switch_prob=0.5,
model='HST_8_384',
nb_classes=1000,
num_workers=10,
output_dir='./output_dir_cml_spikformer',
pin_mem=True,
rank=0,
recount=1,
remode='pixel',
reprob=0.25,
resplit=False,
resume='',
seed=0,
smoothing=0.1,
start_epoch=0,
time_step=4,
warmup_epochs=5,
weight_decay=0.05,
world_size=8)
[10:57:56.778104] Dataset ImageFolder
    Number of datapoints: 1281167
    Root location: /dataset/ImageNet2012/train
    StandardTransform
Transform: Compose(
               RandomResizedCropAndInterpolation(size=(224, 224), scale=(0.08, 1.0), ratio=(0.75, 1.3333), interpolation=PIL.Image.BICUBIC)
               RandomHorizontalFlip(p=0.5)
               <timm.data.auto_augment.RandAugment object at 0x7fbcbf63dc70>
               ToTensor()
               Normalize(mean=tensor([0.4850, 0.4560, 0.4060]), std=tensor([0.2290, 0.2240, 0.2250]))
               <timm.data.random_erasing.RandomErasing object at 0x7fbcbf63df10>
           )
[10:57:57.880839] Dataset ImageFolder
    Number of datapoints: 50000
    Root location: /dataset/ImageNet2012/val
    StandardTransform
Transform: Compose(
               Resize(size=256, interpolation=bicubic, max_size=None, antialias=None)
               CenterCrop(size=(224, 224))
               ToTensor()
               Normalize(mean=(0.485, 0.456, 0.406), std=(0.229, 0.224, 0.225))
           )
[10:57:57.881031] Sampler_train = <torch.utils.data.distributed.DistributedSampler object at 0x7fbcbf63d9d0>
[10:57:58.462166] Model = Spiking_vit(
  (patch_embed1): PatchEmbedInit(
    (proj_conv): Conv2d(3, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
    (proj_bn): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (proj_maxpool): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)
    (proj_lif): MultiStepLIFNode(
      v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
      (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
    )
    (proj1_conv): Conv2d(48, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
    (proj1_bn): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (proj1_maxpool): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)
    (proj1_lif): MultiStepLIFNode(
      v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
      (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
    )
    (proj2_conv): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
    (proj2_bn): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (proj2_lif): MultiStepLIFNode(
      v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
      (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
    )
    (proj_res_conv): Conv2d(48, 96, kernel_size=(1, 1), stride=(2, 2), bias=False)
    (proj_res_bn): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (proj_res_lif): MultiStepLIFNode(
      v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
      (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
    )
  )
  (patch_embed2): PatchEmbeddingStage(
    (proj3_conv): Conv2d(96, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
    (proj3_bn): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (proj3_maxpool): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)
    (proj3_lif): MultiStepLIFNode(
      v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
      (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
    )
    (proj4_conv): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
    (proj4_bn): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (proj4_lif): MultiStepLIFNode(
      v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
      (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
    )
    (proj_res_conv): Conv2d(96, 192, kernel_size=(1, 1), stride=(2, 2), bias=False)
    (proj_res_bn): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (proj_res_lif): MultiStepLIFNode(
      v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
      (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
    )
  )
  (patch_embed3): PatchEmbeddingStage(
    (proj3_conv): Conv2d(192, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
    (proj3_bn): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (proj3_maxpool): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)
    (proj3_lif): MultiStepLIFNode(
      v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
      (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
    )
    (proj4_conv): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
    (proj4_bn): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (proj4_lif): MultiStepLIFNode(
      v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
      (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
    )
    (proj_res_conv): Conv2d(192, 384, kernel_size=(1, 1), stride=(2, 2), bias=False)
    (proj_res_bn): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (proj_res_lif): MultiStepLIFNode(
      v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
      (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
    )
  )
  (stage1): ModuleList(
    (0): TokenSpikingTransformer(
      (tssa): TokenSpikingSelfAttention(
        (q_conv): Conv1d(96, 96, kernel_size=(1,), stride=(1,), bias=False)
        (q_bn): BatchNorm1d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (q_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (k_conv): Conv1d(96, 96, kernel_size=(1,), stride=(1,), bias=False)
        (k_bn): BatchNorm1d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (k_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (attn_lif): MultiStepLIFNode(
          v_threshold=0.5, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (proj_conv): Conv1d(96, 96, kernel_size=(1,), stride=(1,))
        (proj_bn): BatchNorm1d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (proj_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
      )
      (mlp): MLP(
        (fc1_conv): Conv2d(96, 384, kernel_size=(1, 1), stride=(1, 1))
        (fc1_bn): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (fc1_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (fc2_conv): Conv2d(384, 96, kernel_size=(1, 1), stride=(1, 1))
        (fc2_bn): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (fc2_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
      )
    )
  )
  (stage2): ModuleList(
    (0): TokenSpikingTransformer(
      (tssa): TokenSpikingSelfAttention(
        (q_conv): Conv1d(192, 192, kernel_size=(1,), stride=(1,), bias=False)
        (q_bn): BatchNorm1d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (q_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (k_conv): Conv1d(192, 192, kernel_size=(1,), stride=(1,), bias=False)
        (k_bn): BatchNorm1d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (k_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (attn_lif): MultiStepLIFNode(
          v_threshold=0.5, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (proj_conv): Conv1d(192, 192, kernel_size=(1,), stride=(1,))
        (proj_bn): BatchNorm1d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (proj_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
      )
      (mlp): MLP(
        (fc1_conv): Conv2d(192, 768, kernel_size=(1, 1), stride=(1, 1))
        (fc1_bn): BatchNorm2d(768, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (fc1_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (fc2_conv): Conv2d(768, 192, kernel_size=(1, 1), stride=(1, 1))
        (fc2_bn): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (fc2_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
      )
    )
    (1): TokenSpikingTransformer(
      (tssa): TokenSpikingSelfAttention(
        (q_conv): Conv1d(192, 192, kernel_size=(1,), stride=(1,), bias=False)
        (q_bn): BatchNorm1d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (q_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (k_conv): Conv1d(192, 192, kernel_size=(1,), stride=(1,), bias=False)
        (k_bn): BatchNorm1d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (k_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (attn_lif): MultiStepLIFNode(
          v_threshold=0.5, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (proj_conv): Conv1d(192, 192, kernel_size=(1,), stride=(1,))
        (proj_bn): BatchNorm1d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (proj_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
      )
      (mlp): MLP(
        (fc1_conv): Conv2d(192, 768, kernel_size=(1, 1), stride=(1, 1))
        (fc1_bn): BatchNorm2d(768, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (fc1_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (fc2_conv): Conv2d(768, 192, kernel_size=(1, 1), stride=(1, 1))
        (fc2_bn): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (fc2_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
      )
    )
  )
  (stage3): ModuleList(
    (0): SpikingTransformer(
      (attn): Attention(
        (q_conv): Conv1d(384, 384, kernel_size=(1,), stride=(1,), bias=False)
        (q_bn): BatchNorm1d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (q_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (k_conv): Conv1d(384, 384, kernel_size=(1,), stride=(1,), bias=False)
        (k_bn): BatchNorm1d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (k_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (v_conv): Conv1d(384, 384, kernel_size=(1,), stride=(1,), bias=False)
        (v_bn): BatchNorm1d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (v_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (attn_lif): MultiStepLIFNode(
          v_threshold=0.5, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (proj_conv): Conv1d(384, 384, kernel_size=(1,), stride=(1,))
        (proj_bn): BatchNorm1d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (proj_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (qkv_mp): MaxPool1d(kernel_size=4, stride=4, padding=0, dilation=1, ceil_mode=False)
      )
      (mlp): MLP(
        (fc1_conv): Conv2d(384, 1536, kernel_size=(1, 1), stride=(1, 1))
        (fc1_bn): BatchNorm2d(1536, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (fc1_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (fc2_conv): Conv2d(1536, 384, kernel_size=(1, 1), stride=(1, 1))
        (fc2_bn): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (fc2_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
      )
    )
    (1): SpikingTransformer(
      (attn): Attention(
        (q_conv): Conv1d(384, 384, kernel_size=(1,), stride=(1,), bias=False)
        (q_bn): BatchNorm1d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (q_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (k_conv): Conv1d(384, 384, kernel_size=(1,), stride=(1,), bias=False)
        (k_bn): BatchNorm1d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (k_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (v_conv): Conv1d(384, 384, kernel_size=(1,), stride=(1,), bias=False)
        (v_bn): BatchNorm1d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (v_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (attn_lif): MultiStepLIFNode(
          v_threshold=0.5, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (proj_conv): Conv1d(384, 384, kernel_size=(1,), stride=(1,))
        (proj_bn): BatchNorm1d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (proj_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (qkv_mp): MaxPool1d(kernel_size=4, stride=4, padding=0, dilation=1, ceil_mode=False)
      )
      (mlp): MLP(
        (fc1_conv): Conv2d(384, 1536, kernel_size=(1, 1), stride=(1, 1))
        (fc1_bn): BatchNorm2d(1536, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (fc1_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (fc2_conv): Conv2d(1536, 384, kernel_size=(1, 1), stride=(1, 1))
        (fc2_bn): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (fc2_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
      )
    )
    (2): SpikingTransformer(
      (attn): Attention(
        (q_conv): Conv1d(384, 384, kernel_size=(1,), stride=(1,), bias=False)
        (q_bn): BatchNorm1d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (q_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (k_conv): Conv1d(384, 384, kernel_size=(1,), stride=(1,), bias=False)
        (k_bn): BatchNorm1d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (k_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (v_conv): Conv1d(384, 384, kernel_size=(1,), stride=(1,), bias=False)
        (v_bn): BatchNorm1d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (v_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (attn_lif): MultiStepLIFNode(
          v_threshold=0.5, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (proj_conv): Conv1d(384, 384, kernel_size=(1,), stride=(1,))
        (proj_bn): BatchNorm1d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (proj_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (qkv_mp): MaxPool1d(kernel_size=4, stride=4, padding=0, dilation=1, ceil_mode=False)
      )
      (mlp): MLP(
        (fc1_conv): Conv2d(384, 1536, kernel_size=(1, 1), stride=(1, 1))
        (fc1_bn): BatchNorm2d(1536, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (fc1_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (fc2_conv): Conv2d(1536, 384, kernel_size=(1, 1), stride=(1, 1))
        (fc2_bn): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (fc2_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
      )
    )
    (3): SpikingTransformer(
      (attn): Attention(
        (q_conv): Conv1d(384, 384, kernel_size=(1,), stride=(1,), bias=False)
        (q_bn): BatchNorm1d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (q_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (k_conv): Conv1d(384, 384, kernel_size=(1,), stride=(1,), bias=False)
        (k_bn): BatchNorm1d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (k_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (v_conv): Conv1d(384, 384, kernel_size=(1,), stride=(1,), bias=False)
        (v_bn): BatchNorm1d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (v_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (attn_lif): MultiStepLIFNode(
          v_threshold=0.5, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (proj_conv): Conv1d(384, 384, kernel_size=(1,), stride=(1,))
        (proj_bn): BatchNorm1d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (proj_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (qkv_mp): MaxPool1d(kernel_size=4, stride=4, padding=0, dilation=1, ceil_mode=False)
      )
      (mlp): MLP(
        (fc1_conv): Conv2d(384, 1536, kernel_size=(1, 1), stride=(1, 1))
        (fc1_bn): BatchNorm2d(1536, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (fc1_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (fc2_conv): Conv2d(1536, 384, kernel_size=(1, 1), stride=(1, 1))
        (fc2_bn): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (fc2_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
      )
    )
    (4): SpikingTransformer(
      (attn): Attention(
        (q_conv): Conv1d(384, 384, kernel_size=(1,), stride=(1,), bias=False)
        (q_bn): BatchNorm1d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (q_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (k_conv): Conv1d(384, 384, kernel_size=(1,), stride=(1,), bias=False)
        (k_bn): BatchNorm1d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (k_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (v_conv): Conv1d(384, 384, kernel_size=(1,), stride=(1,), bias=False)
        (v_bn): BatchNorm1d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (v_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (attn_lif): MultiStepLIFNode(
          v_threshold=0.5, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (proj_conv): Conv1d(384, 384, kernel_size=(1,), stride=(1,))
        (proj_bn): BatchNorm1d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (proj_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (qkv_mp): MaxPool1d(kernel_size=4, stride=4, padding=0, dilation=1, ceil_mode=False)
      )
      (mlp): MLP(
        (fc1_conv): Conv2d(384, 1536, kernel_size=(1, 1), stride=(1, 1))
        (fc1_bn): BatchNorm2d(1536, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (fc1_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (fc2_conv): Conv2d(1536, 384, kernel_size=(1, 1), stride=(1, 1))
        (fc2_bn): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (fc2_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
      )
    )
    (5): SpikingTransformer(
      (attn): Attention(
        (q_conv): Conv1d(384, 384, kernel_size=(1,), stride=(1,), bias=False)
        (q_bn): BatchNorm1d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (q_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (k_conv): Conv1d(384, 384, kernel_size=(1,), stride=(1,), bias=False)
        (k_bn): BatchNorm1d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (k_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (v_conv): Conv1d(384, 384, kernel_size=(1,), stride=(1,), bias=False)
        (v_bn): BatchNorm1d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (v_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (attn_lif): MultiStepLIFNode(
          v_threshold=0.5, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (proj_conv): Conv1d(384, 384, kernel_size=(1,), stride=(1,))
        (proj_bn): BatchNorm1d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (proj_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (qkv_mp): MaxPool1d(kernel_size=4, stride=4, padding=0, dilation=1, ceil_mode=False)
      )
      (mlp): MLP(
        (fc1_conv): Conv2d(384, 1536, kernel_size=(1, 1), stride=(1, 1))
        (fc1_bn): BatchNorm2d(1536, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (fc1_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (fc2_conv): Conv2d(1536, 384, kernel_size=(1, 1), stride=(1, 1))
        (fc2_bn): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (fc2_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
      )
    )
    (6): SpikingTransformer(
      (attn): Attention(
        (q_conv): Conv1d(384, 384, kernel_size=(1,), stride=(1,), bias=False)
        (q_bn): BatchNorm1d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (q_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (k_conv): Conv1d(384, 384, kernel_size=(1,), stride=(1,), bias=False)
        (k_bn): BatchNorm1d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (k_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (v_conv): Conv1d(384, 384, kernel_size=(1,), stride=(1,), bias=False)
        (v_bn): BatchNorm1d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (v_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (attn_lif): MultiStepLIFNode(
          v_threshold=0.5, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (proj_conv): Conv1d(384, 384, kernel_size=(1,), stride=(1,))
        (proj_bn): BatchNorm1d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (proj_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (qkv_mp): MaxPool1d(kernel_size=4, stride=4, padding=0, dilation=1, ceil_mode=False)
      )
      (mlp): MLP(
        (fc1_conv): Conv2d(384, 1536, kernel_size=(1, 1), stride=(1, 1))
        (fc1_bn): BatchNorm2d(1536, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (fc1_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
        (fc2_conv): Conv2d(1536, 384, kernel_size=(1, 1), stride=(1, 1))
        (fc2_bn): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (fc2_lif): MultiStepLIFNode(
          v_threshold=1.0, v_reset=0.0, detach_reset=True, tau=2.0, backend=cupy
          (surrogate_function): Sigmoid(alpha=4.0, spiking=True)
        )
      )
    )
  )
  (head): Linear(in_features=384, out_features=1000, bias=True)
)
[10:57:58.462223] number of params (M): 16.47
[10:57:58.462240] base lr: 6.00e-04
[10:57:58.462250] actual lr: 1.20e-03
[10:57:58.462258] accumulate grad iterations: 2
[10:57:58.462266] effective batch size: 512
[10:57:58.643850] criterion = LabelSmoothingCrossEntropy()
[10:57:58.643898] Start training for 200 epochs
[10:57:58.644958] log_dir: ./output_dir_cml_spikformer
[W reducer.cpp:1251] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters in the forward pass. This flag results in an extra traversal of the autograd graph every iteration,  which can adversely affect performance. If your model indeed never has any unused parameters in the forward pass, consider turning this flag off. Note that this warning may be a false positive if your model has flow control causing later iterations to have unused parameters. (function operator())
[W reducer.cpp:1251] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters in the forward pass. This flag results in an extra traversal of the autograd graph every iteration,  which can adversely affect performance. If your model indeed never has any unused parameters in the forward pass, consider turning this flag off. Note that this warning may be a false positive if your model has flow control causing later iterations to have unused parameters. (function operator())
[W reducer.cpp:1251] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters in the forward pass. This flag results in an extra traversal of the autograd graph every iteration,  which can adversely affect performance. If your model indeed never has any unused parameters in the forward pass, consider turning this flag off. Note that this warning may be a false positive if your model has flow control causing later iterations to have unused parameters. (function operator())
[W reducer.cpp:1251] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters in the forward pass. This flag results in an extra traversal of the autograd graph every iteration,  which can adversely affect performance. If your model indeed never has any unused parameters in the forward pass, consider turning this flag off. Note that this warning may be a false positive if your model has flow control causing later iterations to have unused parameters. (function operator())
[W reducer.cpp:1251] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters in the forward pass. This flag results in an extra traversal of the autograd graph every iteration,  which can adversely affect performance. If your model indeed never has any unused parameters in the forward pass, consider turning this flag off. Note that this warning may be a false positive if your model has flow control causing later iterations to have unused parameters. (function operator())
[W reducer.cpp:1251] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters in the forward pass. This flag results in an extra traversal of the autograd graph every iteration,  which can adversely affect performance. If your model indeed never has any unused parameters in the forward pass, consider turning this flag off. Note that this warning may be a false positive if your model has flow control causing later iterations to have unused parameters. (function operator())
[W reducer.cpp:1251] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters in the forward pass. This flag results in an extra traversal of the autograd graph every iteration,  which can adversely affect performance. If your model indeed never has any unused parameters in the forward pass, consider turning this flag off. Note that this warning may be a false positive if your model has flow control causing later iterations to have unused parameters. (function operator())
[W reducer.cpp:1251] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters in the forward pass. This flag results in an extra traversal of the autograd graph every iteration,  which can adversely affect performance. If your model indeed never has any unused parameters in the forward pass, consider turning this flag off. Note that this warning may be a false positive if your model has flow control causing later iterations to have unused parameters. (function operator())
[10:58:39.153690] Epoch: [0]  [   0/5004]  eta: 2 days, 8:11:00  lr: 0.000000  loss: 6.9099 (6.9099)  time: 40.4197  data: 2.2660  max mem: 17867
[11:10:10.988714] Epoch: [0]  [2000/5004]  eta: 0:18:19  lr: 0.000096  loss: 6.6493 (6.8347)  time: 0.3438  data: 0.0002  max mem: 17867
[11:21:40.897336] Epoch: [0]  [4000/5004]  eta: 0:05:56  lr: 0.000192  loss: 6.2546 (6.6452)  time: 0.3435  data: 0.0002  max mem: 17867
[11:27:27.079603] Epoch: [0]  [5003/5004]  eta: 0:00:00  lr: 0.000240  loss: 6.1045 (6.5492)  time: 0.3423  data: 0.0005  max mem: 17867
[11:27:27.408769] Epoch: [0] Total time: 0:29:28 (0.3535 s / it)
[11:27:27.417850] Averaged stats: lr: 0.000240  loss: 6.1045 (6.5492)
[11:27:28.759501] Test:  [   0/1563]  eta: 0:34:51  loss: 6.2096 (6.2096)  acc1: 9.3750 (9.3750)  acc5: 18.7500 (18.7500)  time: 1.3384  data: 0.9835  max mem: 17867
[11:28:33.820166] Test:  [ 500/1563]  eta: 0:02:20  loss: 5.3346 (5.8837)  acc1: 0.0000 (2.2206)  acc5: 0.0000 (9.2939)  time: 0.1300  data: 0.0002  max mem: 17867
[11:29:38.817968] Test:  [1000/1563]  eta: 0:01:13  loss: 5.7247 (5.7241)  acc1: 0.0000 (3.1156)  acc5: 3.1250 (10.9828)  time: 0.1300  data: 0.0002  max mem: 17867
[11:30:43.811773] Test:  [1500/1563]  eta: 0:00:08  loss: 4.5013 (5.6138)  acc1: 3.1250 (3.6372)  acc5: 18.7500 (12.3210)  time: 0.1300  data: 0.0004  max mem: 17867
[11:30:52.389402] Test:  [1562/1563]  eta: 0:00:00  loss: 5.4401 (5.5925)  acc1: 0.0000 (3.8280)  acc5: 6.2500 (12.6880)  time: 0.1559  data: 0.0001  max mem: 17867
[11:30:52.523033] Test: Total time: 0:03:25 (0.1312 s / it)
[11:30:52.529707] * Acc@1 3.828 Acc@5 12.688 loss 5.593
[11:30:52.529888] Accuracy of the network on the 50000 test images: 3.8%
[11:30:52.529919] Max accuracy: 3.83%
[11:30:52.596555] log_dir: ./output_dir_cml_spikformer
[11:30:54.511796] Epoch: [1]  [   0/5004]  eta: 2:39:40  lr: 0.000240  loss: 5.8621 (5.8621)  time: 1.9145  data: 1.0186  max mem: 17867
[11:42:25.014347] Epoch: [1]  [2000/5004]  eta: 0:17:19  lr: 0.000336  loss: 5.7659 (5.9320)  time: 0.3447  data: 0.0002  max mem: 17867
[11:53:55.192830] Epoch: [1]  [4000/5004]  eta: 0:05:46  lr: 0.000432  loss: 5.4404 (5.8009)  time: 0.3427  data: 0.0002  max mem: 17867
[11:59:41.518006] Epoch: [1]  [5003/5004]  eta: 0:00:00  lr: 0.000480  loss: 5.5087 (5.7362)  time: 0.3434  data: 0.0006  max mem: 17867
[11:59:41.846489] Epoch: [1] Total time: 0:28:49 (0.3456 s / it)
[11:59:41.851955] Averaged stats: lr: 0.000480  loss: 5.5087 (5.7359)
[11:59:42.793471] Test:  [   0/1563]  eta: 0:24:24  loss: 3.8763 (3.8763)  acc1: 46.8750 (46.8750)  acc5: 56.2500 (56.2500)  time: 0.9368  data: 0.7721  max mem: 17867
[12:00:48.487783] Test:  [ 500/1563]  eta: 0:02:21  loss: 4.3856 (4.1490)  acc1: 6.2500 (15.2570)  acc5: 31.2500 (38.1549)  time: 0.1299  data: 0.0002  max mem: 17867
[12:01:53.579781] Test:  [1000/1563]  eta: 0:01:14  loss: 4.2143 (4.2583)  acc1: 9.3750 (15.6718)  acc5: 40.6250 (37.3252)  time: 0.1299  data: 0.0002  max mem: 17867
[12:02:58.697454] Test:  [1500/1563]  eta: 0:00:08  loss: 3.3110 (4.3244)  acc1: 25.0000 (14.9692)  acc5: 53.1250 (35.8698)  time: 0.1299  data: 0.0002  max mem: 17867
[12:03:06.681297] Test:  [1562/1563]  eta: 0:00:00  loss: 3.8158 (4.3030)  acc1: 18.7500 (15.3420)  acc5: 46.8750 (36.3720)  time: 0.1262  data: 0.0001  max mem: 17867
[12:03:06.747464] Test: Total time: 0:03:24 (0.1311 s / it)
[12:03:06.955705] * Acc@1 15.342 Acc@5 36.372 loss 4.303
[12:03:06.955848] Accuracy of the network on the 50000 test images: 15.3%
[12:03:06.955870] Max accuracy: 15.34%
[12:03:06.999122] log_dir: ./output_dir_cml_spikformer
[12:03:09.460104] Epoch: [2]  [   0/5004]  eta: 3:25:10  lr: 0.000480  loss: 5.4414 (5.4414)  time: 2.4600  data: 1.8033  max mem: 17867
[12:14:40.802963] Epoch: [2]  [2000/5004]  eta: 0:17:21  lr: 0.000576  loss: 5.2722 (5.3005)  time: 0.3453  data: 0.0002  max mem: 17867
[12:26:11.331561] Epoch: [2]  [4000/5004]  eta: 0:05:47  lr: 0.000672  loss: 4.9452 (5.2196)  time: 0.3464  data: 0.0002  max mem: 17867
[12:31:57.804605] Epoch: [2]  [5003/5004]  eta: 0:00:00  lr: 0.000720  loss: 4.9029 (5.1827)  time: 0.3473  data: 0.0006  max mem: 17867
[12:31:58.145273] Epoch: [2] Total time: 0:28:51 (0.3460 s / it)
[12:31:58.163676] Averaged stats: lr: 0.000720  loss: 4.9029 (5.1834)
[12:31:59.140616] Test:  [   0/1563]  eta: 0:25:21  loss: 2.8175 (2.8175)  acc1: 46.8750 (46.8750)  acc5: 71.8750 (71.8750)  time: 0.9736  data: 0.8288  max mem: 17867
[12:33:04.121401] Test:  [ 500/1563]  eta: 0:02:19  loss: 3.7996 (3.4713)  acc1: 9.3750 (22.8543)  acc5: 43.7500 (50.9232)  time: 0.1299  data: 0.0002  max mem: 17867
[12:34:09.247441] Test:  [1000/1563]  eta: 0:01:13  loss: 3.7093 (3.6905)  acc1: 18.7500 (22.1310)  acc5: 50.0000 (47.5212)  time: 0.1299  data: 0.0002  max mem: 17867
[12:35:14.210636] Test:  [1500/1563]  eta: 0:00:08  loss: 3.0465 (3.8146)  acc1: 34.3750 (21.6043)  acc5: 62.5000 (45.3989)  time: 0.1299  data: 0.0002  max mem: 17867
[12:35:22.192992] Test:  [1562/1563]  eta: 0:00:00  loss: 2.5409 (3.7918)  acc1: 40.6250 (22.0580)  acc5: 68.7500 (45.8480)  time: 0.1262  data: 0.0001  max mem: 17867
[12:35:22.272969] Test: Total time: 0:03:24 (0.1306 s / it)
[12:35:22.381564] * Acc@1 22.058 Acc@5 45.848 loss 3.792
[12:35:22.381699] Accuracy of the network on the 50000 test images: 22.1%
[12:35:22.381721] Max accuracy: 22.06%
[12:35:22.405232] log_dir: ./output_dir_cml_spikformer
[12:35:23.859990] Epoch: [3]  [   0/5004]  eta: 2:01:15  lr: 0.000720  loss: 4.7863 (4.7863)  time: 1.4540  data: 0.9307  max mem: 17867
[12:46:59.011415] Epoch: [3]  [2000/5004]  eta: 0:17:25  lr: 0.000816  loss: 4.8892 (4.9107)  time: 0.3488  data: 0.0002  max mem: 17867
[12:58:32.410840] Epoch: [3]  [4000/5004]  eta: 0:05:48  lr: 0.000912  loss: 4.6710 (4.8391)  time: 0.3434  data: 0.0002  max mem: 17867
[13:04:19.727205] Epoch: [3]  [5003/5004]  eta: 0:00:00  lr: 0.000960  loss: 4.6850 (4.8094)  time: 0.3455  data: 0.0006  max mem: 17867
[13:04:20.074701] Epoch: [3] Total time: 0:28:57 (0.3473 s / it)
[13:04:20.083978] Averaged stats: lr: 0.000960  loss: 4.6850 (4.8089)
[13:04:21.072490] Test:  [   0/1563]  eta: 0:25:39  loss: 2.4880 (2.4880)  acc1: 40.6250 (40.6250)  acc5: 75.0000 (75.0000)  time: 0.9850  data: 0.8050  max mem: 17867
[13:05:26.036705] Test:  [ 500/1563]  eta: 0:02:19  loss: 2.9443 (2.9067)  acc1: 28.1250 (32.6347)  acc5: 62.5000 (62.8368)  time: 0.1298  data: 0.0002  max mem: 17867
[13:06:30.962352] Test:  [1000/1563]  eta: 0:01:13  loss: 2.8278 (3.0731)  acc1: 28.1250 (31.7932)  acc5: 68.7500 (60.2835)  time: 0.1298  data: 0.0002  max mem: 17867
[13:07:35.918663] Test:  [1500/1563]  eta: 0:00:08  loss: 2.1572 (3.1956)  acc1: 50.0000 (30.6691)  acc5: 75.0000 (58.2049)  time: 0.1298  data: 0.0002  max mem: 17867
[13:07:43.984929] Test:  [1562/1563]  eta: 0:00:00  loss: 1.7428 (3.1797)  acc1: 50.0000 (31.0620)  acc5: 81.2500 (58.5280)  time: 0.1304  data: 0.0001  max mem: 17867
[13:07:44.059307] Test: Total time: 0:03:23 (0.1305 s / it)
[13:07:44.492192] * Acc@1 31.062 Acc@5 58.528 loss 3.180
[13:07:44.492359] Accuracy of the network on the 50000 test images: 31.1%
[13:07:44.492382] Max accuracy: 31.06%
[13:07:44.498757] log_dir: ./output_dir_cml_spikformer
[13:07:45.993835] Epoch: [4]  [   0/5004]  eta: 2:04:37  lr: 0.000960  loss: 4.5506 (4.5506)  time: 1.4944  data: 0.9963  max mem: 17867
[13:19:19.028843] Epoch: [4]  [2000/5004]  eta: 0:17:22  lr: 0.001056  loss: 4.5681 (4.6000)  time: 0.3433  data: 0.0002  max mem: 17867
[13:30:51.120650] Epoch: [4]  [4000/5004]  eta: 0:05:47  lr: 0.001152  loss: 4.4422 (4.5619)  time: 0.3478  data: 0.0002  max mem: 17867
[13:36:37.822897] Epoch: [4]  [5003/5004]  eta: 0:00:00  lr: 0.001200  loss: 4.4624 (4.5416)  time: 0.3425  data: 0.0006  max mem: 17867
[13:36:38.122648] Epoch: [4] Total time: 0:28:53 (0.3464 s / it)
[13:36:38.128825] Averaged stats: lr: 0.001200  loss: 4.4624 (4.5414)
[13:36:39.208234] Test:  [   0/1563]  eta: 0:28:01  loss: 0.9551 (0.9551)  acc1: 84.3750 (84.3750)  acc5: 93.7500 (93.7500)  time: 1.0759  data: 0.8881  max mem: 17867
[13:37:44.196836] Test:  [ 500/1563]  eta: 0:02:20  loss: 2.7329 (2.5014)  acc1: 31.2500 (40.1260)  acc5: 65.6250 (70.6712)  time: 0.1298  data: 0.0002  max mem: 17867
[13:38:49.148095] Test:  [1000/1563]  eta: 0:01:13  loss: 2.8806 (2.7340)  acc1: 34.3750 (37.6155)  acc5: 65.6250 (66.5366)  time: 0.1298  data: 0.0002  max mem: 17867
[13:39:54.086434] Test:  [1500/1563]  eta: 0:00:08  loss: 2.0434 (2.8961)  acc1: 53.1250 (35.3098)  acc5: 78.1250 (63.6472)  time: 0.1298  data: 0.0002  max mem: 17867
[13:40:02.066486] Test:  [1562/1563]  eta: 0:00:00  loss: 1.5459 (2.8802)  acc1: 62.5000 (35.6920)  acc5: 84.3750 (63.9440)  time: 0.1261  data: 0.0001  max mem: 17867
[13:40:02.122798] Test: Total time: 0:03:23 (0.1305 s / it)
[13:40:02.470528] * Acc@1 35.692 Acc@5 63.944 loss 2.880
[13:40:02.470676] Accuracy of the network on the 50000 test images: 35.7%
[13:40:02.470697] Max accuracy: 35.69%
[13:40:02.511744] log_dir: ./output_dir_cml_spikformer
[13:40:04.126888] Epoch: [5]  [   0/5004]  eta: 2:14:38  lr: 0.001200  loss: 4.6272 (4.6272)  time: 1.6143  data: 0.8835  max mem: 17867
[13:51:38.891500] Epoch: [5]  [2000/5004]  eta: 0:17:25  lr: 0.001200  loss: 4.3562 (4.3941)  time: 0.3500  data: 0.0002  max mem: 17867
[14:03:12.788683] Epoch: [5]  [4000/5004]  eta: 0:05:48  lr: 0.001200  loss: 4.3834 (4.3519)  time: 0.3468  data: 0.0002  max mem: 17867
[14:09:01.013597] Epoch: [5]  [5003/5004]  eta: 0:00:00  lr: 0.001200  loss: 4.1961 (4.3301)  time: 0.3422  data: 0.0006  max mem: 17867
[14:09:01.343022] Epoch: [5] Total time: 0:28:58 (0.3475 s / it)
[14:09:01.347967] Averaged stats: lr: 0.001200  loss: 4.1961 (4.3284)
[14:09:02.313095] Test:  [   0/1563]  eta: 0:25:03  loss: 0.9898 (0.9898)  acc1: 84.3750 (84.3750)  acc5: 96.8750 (96.8750)  time: 0.9617  data: 0.8224  max mem: 17867
[14:10:07.383421] Test:  [ 500/1563]  eta: 0:02:20  loss: 2.3930 (2.2161)  acc1: 40.6250 (45.7834)  acc5: 75.0000 (76.5344)  time: 0.1298  data: 0.0002  max mem: 17867
[14:11:12.351041] Test:  [1000/1563]  eta: 0:01:13  loss: 2.7867 (2.4726)  acc1: 37.5000 (43.1943)  acc5: 71.8750 (71.9624)  time: 0.1298  data: 0.0002  max mem: 17867
[14:12:17.281441] Test:  [1500/1563]  eta: 0:00:08  loss: 1.8551 (2.6071)  acc1: 59.3750 (41.7139)  acc5: 81.2500 (69.5120)  time: 0.1298  data: 0.0002  max mem: 17867
[14:12:25.256767] Test:  [1562/1563]  eta: 0:00:00  loss: 1.2842 (2.5926)  acc1: 65.6250 (42.0200)  acc5: 90.6250 (69.7500)  time: 0.1261  data: 0.0001  max mem: 17867
[14:12:25.321624] Test: Total time: 0:03:23 (0.1305 s / it)
[14:12:25.567968] * Acc@1 42.020 Acc@5 69.750 loss 2.593
[14:12:25.568136] Accuracy of the network on the 50000 test images: 42.0%
[14:12:25.568157] Max accuracy: 42.02%
[14:12:25.590377] log_dir: ./output_dir_cml_spikformer
[14:12:27.236217] Epoch: [6]  [   0/5004]  eta: 2:17:11  lr: 0.001200  loss: 4.5830 (4.5830)  time: 1.6449  data: 1.3156  max mem: 17867
[14:24:03.719475] Epoch: [6]  [2000/5004]  eta: 0:17:27  lr: 0.001200  loss: 4.1657 (4.1883)  time: 0.3483  data: 0.0002  max mem: 17867
[14:35:38.858778] Epoch: [6]  [4000/5004]  eta: 0:05:49  lr: 0.001200  loss: 4.0418 (4.1678)  time: 0.3536  data: 0.0002  max mem: 17867
[14:41:26.937419] Epoch: [6]  [5003/5004]  eta: 0:00:00  lr: 0.001200  loss: 3.9995 (4.1519)  time: 0.3425  data: 0.0006  max mem: 17867
[14:41:27.242504] Epoch: [6] Total time: 0:29:01 (0.3481 s / it)
[14:41:27.245132] Averaged stats: lr: 0.001200  loss: 3.9995 (4.1494)
[14:41:28.618214] Test:  [   0/1563]  eta: 0:35:40  loss: 0.8925 (0.8925)  acc1: 87.5000 (87.5000)  acc5: 93.7500 (93.7500)  time: 1.3696  data: 1.2282  max mem: 17867
[14:42:33.685092] Test:  [ 500/1563]  eta: 0:02:20  loss: 2.0840 (2.0398)  acc1: 43.7500 (50.5177)  acc5: 81.2500 (79.7779)  time: 0.1299  data: 0.0002  max mem: 17867
[14:43:38.681488] Test:  [1000/1563]  eta: 0:01:13  loss: 2.6293 (2.2610)  acc1: 37.5000 (47.8334)  acc5: 78.1250 (75.3184)  time: 0.1299  data: 0.0002  max mem: 17867
[14:44:43.669063] Test:  [1500/1563]  eta: 0:00:08  loss: 1.5322 (2.4014)  acc1: 62.5000 (45.4364)  acc5: 87.5000 (72.6786)  time: 0.1299  data: 0.0002  max mem: 17867
[14:44:51.655262] Test:  [1562/1563]  eta: 0:00:00  loss: 1.5083 (2.3971)  acc1: 65.6250 (45.5040)  acc5: 87.5000 (72.7620)  time: 0.1262  data: 0.0001  max mem: 17867
[14:44:51.733024] Test: Total time: 0:03:24 (0.1308 s / it)
[14:44:51.957637] * Acc@1 45.504 Acc@5 72.762 loss 2.397
[14:44:51.957810] Accuracy of the network on the 50000 test images: 45.5%
[14:44:51.957832] Max accuracy: 45.50%
[14:44:51.964266] log_dir: ./output_dir_cml_spikformer
[14:44:53.594685] Epoch: [7]  [   0/5004]  eta: 2:15:52  lr: 0.001200  loss: 3.8549 (3.8549)  time: 1.6293  data: 0.9937  max mem: 17867
[14:56:27.958185] Epoch: [7]  [2000/5004]  eta: 0:17:24  lr: 0.001200  loss: 4.0950 (4.0564)  time: 0.3492  data: 0.0002  max mem: 17867
[15:08:03.523814] Epoch: [7]  [4000/5004]  eta: 0:05:49  lr: 0.001199  loss: 4.0588 (4.0291)  time: 0.3450  data: 0.0002  max mem: 17867
[15:13:51.651338] Epoch: [7]  [5003/5004]  eta: 0:00:00  lr: 0.001199  loss: 4.0260 (4.0184)  time: 0.3423  data: 0.0006  max mem: 17867
[15:13:51.981644] Epoch: [7] Total time: 0:29:00 (0.3477 s / it)
[15:13:51.982423] Averaged stats: lr: 0.001199  loss: 4.0260 (4.0264)
[15:13:53.001954] Test:  [   0/1563]  eta: 0:26:25  loss: 0.8806 (0.8806)  acc1: 90.6250 (90.6250)  acc5: 93.7500 (93.7500)  time: 1.0146  data: 0.8746  max mem: 17867
[15:14:58.314058] Test:  [ 500/1563]  eta: 0:02:20  loss: 2.1825 (2.0563)  acc1: 37.5000 (50.2495)  acc5: 81.2500 (79.3975)  time: 0.1314  data: 0.0002  max mem: 17867
[15:16:03.597169] Test:  [1000/1563]  eta: 0:01:14  loss: 2.3347 (2.2427)  acc1: 40.6250 (48.0176)  acc5: 71.8750 (75.9615)  time: 0.1301  data: 0.0002  max mem: 17867
[15:17:08.561901] Test:  [1500/1563]  eta: 0:00:08  loss: 1.5303 (2.3467)  acc1: 65.6250 (46.4773)  acc5: 87.5000 (73.9257)  time: 0.1299  data: 0.0002  max mem: 17867
[15:17:16.545536] Test:  [1562/1563]  eta: 0:00:00  loss: 1.3576 (2.3363)  acc1: 71.8750 (46.7440)  acc5: 90.6250 (74.0540)  time: 0.1262  data: 0.0001  max mem: 17867
[15:17:16.630858] Test: Total time: 0:03:24 (0.1309 s / it)
[15:17:16.632095] * Acc@1 46.744 Acc@5 74.054 loss 2.336
[15:17:16.632229] Accuracy of the network on the 50000 test images: 46.7%
[15:17:16.632251] Max accuracy: 46.74%
[15:17:16.687764] log_dir: ./output_dir_cml_spikformer
[15:17:18.005808] Epoch: [8]  [   0/5004]  eta: 1:49:52  lr: 0.001199  loss: 4.2855 (4.2855)  time: 1.3174  data: 0.9216  max mem: 17867
[15:28:55.610695] Epoch: [8]  [2000/5004]  eta: 0:17:29  lr: 0.001199  loss: 3.9820 (3.9420)  time: 0.3478  data: 0.0002  max mem: 17867
[15:40:29.860904] Epoch: [8]  [4000/5004]  eta: 0:05:49  lr: 0.001199  loss: 3.9272 (3.9379)  time: 0.3434  data: 0.0002  max mem: 17867
[15:46:18.754359] Epoch: [8]  [5003/5004]  eta: 0:00:00  lr: 0.001199  loss: 3.8627 (3.9298)  time: 0.3464  data: 0.0011  max mem: 17867
[15:46:19.065578] Epoch: [8] Total time: 0:29:02 (0.3482 s / it)
[15:46:19.068539] Averaged stats: lr: 0.001199  loss: 3.8627 (3.9259)
[15:46:20.037941] Test:  [   0/1563]  eta: 0:25:09  loss: 1.0165 (1.0165)  acc1: 81.2500 (81.2500)  acc5: 90.6250 (90.6250)  time: 0.9659  data: 0.8191  max mem: 17867
[15:47:25.121673] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.8417 (1.7982)  acc1: 50.0000 (55.9194)  acc5: 81.2500 (83.5329)  time: 0.1300  data: 0.0002  max mem: 17867
[15:48:30.305837] Test:  [1000/1563]  eta: 0:01:13  loss: 2.2496 (2.0075)  acc1: 40.6250 (52.7223)  acc5: 78.1250 (79.4237)  time: 0.1299  data: 0.0002  max mem: 17867
[15:49:35.341779] Test:  [1500/1563]  eta: 0:00:08  loss: 1.5360 (2.1306)  acc1: 65.6250 (50.6267)  acc5: 87.5000 (77.3901)  time: 0.1300  data: 0.0002  max mem: 17867
[15:49:43.326718] Test:  [1562/1563]  eta: 0:00:00  loss: 1.2533 (2.1289)  acc1: 75.0000 (50.7140)  acc5: 87.5000 (77.4180)  time: 0.1262  data: 0.0001  max mem: 17867
[15:49:43.409175] Test: Total time: 0:03:24 (0.1307 s / it)
[15:49:43.585669] * Acc@1 50.714 Acc@5 77.418 loss 2.129
[15:49:43.585853] Accuracy of the network on the 50000 test images: 50.7%
[15:49:43.585875] Max accuracy: 50.71%
[15:49:43.616803] log_dir: ./output_dir_cml_spikformer
[15:49:45.049901] Epoch: [9]  [   0/5004]  eta: 1:59:26  lr: 0.001199  loss: 4.4336 (4.4336)  time: 1.4321  data: 0.9694  max mem: 17867
[16:01:20.548773] Epoch: [9]  [2000/5004]  eta: 0:17:26  lr: 0.001198  loss: 3.7216 (3.8672)  time: 0.3436  data: 0.0002  max mem: 17867
[16:12:56.440457] Epoch: [9]  [4000/5004]  eta: 0:05:49  lr: 0.001198  loss: 3.8737 (3.8582)  time: 0.3503  data: 0.0002  max mem: 17867
[16:18:44.813032] Epoch: [9]  [5003/5004]  eta: 0:00:00  lr: 0.001198  loss: 3.9448 (3.8547)  time: 0.3423  data: 0.0011  max mem: 17867
[16:18:45.123351] Epoch: [9] Total time: 0:29:01 (0.3480 s / it)
[16:18:45.124181] Averaged stats: lr: 0.001198  loss: 3.9448 (3.8532)
[16:18:46.159668] Test:  [   0/1563]  eta: 0:26:52  loss: 1.0935 (1.0935)  acc1: 78.1250 (78.1250)  acc5: 93.7500 (93.7500)  time: 1.0318  data: 0.8889  max mem: 17867
[16:19:51.270874] Test:  [ 500/1563]  eta: 0:02:20  loss: 2.2310 (1.8121)  acc1: 40.6250 (55.8196)  acc5: 75.0000 (83.4082)  time: 0.1299  data: 0.0002  max mem: 17867
[16:20:56.240408] Test:  [1000/1563]  eta: 0:01:13  loss: 2.5166 (2.0655)  acc1: 37.5000 (52.3945)  acc5: 78.1250 (78.9148)  time: 0.1299  data: 0.0002  max mem: 17867
[16:22:01.230752] Test:  [1500/1563]  eta: 0:00:08  loss: 1.3930 (2.1840)  acc1: 71.8750 (50.5621)  acc5: 87.5000 (76.8633)  time: 0.1298  data: 0.0002  max mem: 17867
[16:22:09.209215] Test:  [1562/1563]  eta: 0:00:00  loss: 1.2590 (2.1853)  acc1: 65.6250 (50.5860)  acc5: 90.6250 (76.8380)  time: 0.1261  data: 0.0001  max mem: 17867
[16:22:09.272870] Test: Total time: 0:03:24 (0.1306 s / it)
[16:22:09.428487] * Acc@1 50.586 Acc@5 76.838 loss 2.185
[16:22:09.428623] Accuracy of the network on the 50000 test images: 50.6%
[16:22:09.428645] Max accuracy: 50.71%
[16:22:09.468713] log_dir: ./output_dir_cml_spikformer
[16:22:11.145791] Epoch: [10]  [   0/5004]  eta: 2:19:49  lr: 0.001198  loss: 3.6226 (3.6226)  time: 1.6765  data: 0.9968  max mem: 17867
[16:33:47.245588] Epoch: [10]  [2000/5004]  eta: 0:17:27  lr: 0.001198  loss: 3.8119 (3.8107)  time: 0.3530  data: 0.0002  max mem: 17867
[16:45:22.552580] Epoch: [10]  [4000/5004]  eta: 0:05:49  lr: 0.001197  loss: 3.7378 (3.8033)  time: 0.3440  data: 0.0002  max mem: 17867
[16:51:10.631724] Epoch: [10]  [5003/5004]  eta: 0:00:00  lr: 0.001197  loss: 3.6730 (3.7955)  time: 0.3424  data: 0.0006  max mem: 17867
[16:51:10.943730] Epoch: [10] Total time: 0:29:01 (0.3480 s / it)
[16:51:10.944369] Averaged stats: lr: 0.001197  loss: 3.6730 (3.7907)
[16:51:12.104821] Test:  [   0/1563]  eta: 0:30:08  loss: 1.0937 (1.0937)  acc1: 75.0000 (75.0000)  acc5: 93.7500 (93.7500)  time: 1.1571  data: 0.9686  max mem: 17867
[16:52:17.205395] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.8133 (1.7503)  acc1: 56.2500 (56.8738)  acc5: 84.3750 (84.6058)  time: 0.1301  data: 0.0002  max mem: 17867
[16:53:22.216524] Test:  [1000/1563]  eta: 0:01:13  loss: 1.9231 (1.9372)  acc1: 56.2500 (54.1708)  acc5: 78.1250 (81.0627)  time: 0.1299  data: 0.0002  max mem: 17867
[16:54:27.221546] Test:  [1500/1563]  eta: 0:00:08  loss: 1.2358 (2.0506)  acc1: 68.7500 (52.3380)  acc5: 87.5000 (78.8870)  time: 0.1299  data: 0.0002  max mem: 17867
[16:54:35.207609] Test:  [1562/1563]  eta: 0:00:00  loss: 1.0374 (2.0397)  acc1: 78.1250 (52.6540)  acc5: 87.5000 (79.0440)  time: 0.1262  data: 0.0001  max mem: 17867
[16:54:35.293029] Test: Total time: 0:03:24 (0.1307 s / it)
[16:54:35.460372] * Acc@1 52.654 Acc@5 79.044 loss 2.040
[16:54:35.460518] Accuracy of the network on the 50000 test images: 52.7%
[16:54:35.460539] Max accuracy: 52.65%
[16:54:35.475302] log_dir: ./output_dir_cml_spikformer
[16:54:36.925272] Epoch: [11]  [   0/5004]  eta: 2:00:52  lr: 0.001197  loss: 3.8228 (3.8228)  time: 1.4493  data: 0.9865  max mem: 17867
[17:06:12.208323] Epoch: [11]  [2000/5004]  eta: 0:17:25  lr: 0.001197  loss: 3.6336 (3.7432)  time: 0.3468  data: 0.0002  max mem: 17867
[17:17:47.168609] Epoch: [11]  [4000/5004]  eta: 0:05:49  lr: 0.001196  loss: 3.6628 (3.7420)  time: 0.3481  data: 0.0002  max mem: 17867
[17:23:35.603778] Epoch: [11]  [5003/5004]  eta: 0:00:00  lr: 0.001196  loss: 3.6870 (3.7355)  time: 0.3437  data: 0.0006  max mem: 17867
[17:23:35.920473] Epoch: [11] Total time: 0:29:00 (0.3478 s / it)
[17:23:35.930537] Averaged stats: lr: 0.001196  loss: 3.6870 (3.7398)
[17:23:36.888922] Test:  [   0/1563]  eta: 0:24:52  loss: 0.8733 (0.8733)  acc1: 87.5000 (87.5000)  acc5: 90.6250 (90.6250)  time: 0.9549  data: 0.7655  max mem: 17867
[17:24:41.845799] Test:  [ 500/1563]  eta: 0:02:19  loss: 1.5670 (1.6447)  acc1: 53.1250 (59.5621)  acc5: 84.3750 (86.2463)  time: 0.1299  data: 0.0002  max mem: 17867
[17:25:46.823157] Test:  [1000/1563]  eta: 0:01:13  loss: 2.1314 (1.8142)  acc1: 43.7500 (56.7838)  acc5: 84.3750 (82.7797)  time: 0.1298  data: 0.0002  max mem: 17867
[17:26:51.788953] Test:  [1500/1563]  eta: 0:00:08  loss: 1.2954 (1.9499)  acc1: 68.7500 (54.3263)  acc5: 87.5000 (80.3131)  time: 0.1299  data: 0.0002  max mem: 17867
[17:26:59.769809] Test:  [1562/1563]  eta: 0:00:00  loss: 0.9782 (1.9438)  acc1: 81.2500 (54.5100)  acc5: 90.6250 (80.3620)  time: 0.1261  data: 0.0001  max mem: 17867
[17:26:59.841112] Test: Total time: 0:03:23 (0.1305 s / it)
[17:27:00.257193] * Acc@1 54.510 Acc@5 80.362 loss 1.944
[17:27:00.257367] Accuracy of the network on the 50000 test images: 54.5%
[17:27:00.257395] Max accuracy: 54.51%
[17:27:00.264142] log_dir: ./output_dir_cml_spikformer
[17:27:01.703314] Epoch: [12]  [   0/5004]  eta: 1:59:56  lr: 0.001196  loss: 3.8579 (3.8579)  time: 1.4382  data: 0.8554  max mem: 17867
[17:38:36.686671] Epoch: [12]  [2000/5004]  eta: 0:17:25  lr: 0.001196  loss: 3.6540 (3.7047)  time: 0.3466  data: 0.0002  max mem: 17867
[17:50:11.946545] Epoch: [12]  [4000/5004]  eta: 0:05:49  lr: 0.001195  loss: 3.6952 (3.7083)  time: 0.3491  data: 0.0002  max mem: 17867
[17:56:00.201832] Epoch: [12]  [5003/5004]  eta: 0:00:00  lr: 0.001195  loss: 3.6380 (3.7044)  time: 0.3430  data: 0.0006  max mem: 17867
[17:56:00.495799] Epoch: [12] Total time: 0:29:00 (0.3478 s / it)
[17:56:00.527481] Averaged stats: lr: 0.001195  loss: 3.6380 (3.6981)
[17:56:02.124793] Test:  [   0/1563]  eta: 0:41:30  loss: 1.0812 (1.0812)  acc1: 87.5000 (87.5000)  acc5: 90.6250 (90.6250)  time: 1.5936  data: 1.4577  max mem: 17867
[17:57:07.192775] Test:  [ 500/1563]  eta: 0:02:21  loss: 1.7767 (1.6053)  acc1: 56.2500 (61.0217)  acc5: 84.3750 (86.8263)  time: 0.1299  data: 0.0002  max mem: 17867
[17:58:12.175071] Test:  [1000/1563]  eta: 0:01:14  loss: 1.9872 (1.8040)  acc1: 46.8750 (57.5518)  acc5: 84.3750 (83.1450)  time: 0.1299  data: 0.0002  max mem: 17867
[17:59:17.148201] Test:  [1500/1563]  eta: 0:00:08  loss: 1.4984 (1.9509)  acc1: 65.6250 (54.8239)  acc5: 87.5000 (80.4276)  time: 0.1299  data: 0.0002  max mem: 17867
[17:59:25.130525] Test:  [1562/1563]  eta: 0:00:00  loss: 1.1148 (1.9431)  acc1: 78.1250 (54.9760)  acc5: 90.6250 (80.5500)  time: 0.1262  data: 0.0001  max mem: 17867
[17:59:25.196139] Test: Total time: 0:03:24 (0.1309 s / it)
[17:59:25.766478] * Acc@1 54.976 Acc@5 80.550 loss 1.943
[17:59:25.766623] Accuracy of the network on the 50000 test images: 55.0%
[17:59:25.766644] Max accuracy: 54.98%
[17:59:25.813847] log_dir: ./output_dir_cml_spikformer
[17:59:27.353027] Epoch: [13]  [   0/5004]  eta: 2:08:19  lr: 0.001195  loss: 3.6068 (3.6068)  time: 1.5386  data: 0.9212  max mem: 17867
[18:11:02.623872] Epoch: [13]  [2000/5004]  eta: 0:17:26  lr: 0.001195  loss: 3.4917 (3.6543)  time: 0.3551  data: 0.0002  max mem: 17867
[18:22:37.157934] Epoch: [13]  [4000/5004]  eta: 0:05:49  lr: 0.001194  loss: 3.7493 (3.6580)  time: 0.3453  data: 0.0002  max mem: 17867
[18:28:25.497417] Epoch: [13]  [5003/5004]  eta: 0:00:00  lr: 0.001194  loss: 3.5894 (3.6585)  time: 0.3466  data: 0.0011  max mem: 17867
[18:28:25.817776] Epoch: [13] Total time: 0:29:00 (0.3477 s / it)
[18:28:25.818517] Averaged stats: lr: 0.001194  loss: 3.5894 (3.6635)
[18:28:26.858740] Test:  [   0/1563]  eta: 0:27:00  loss: 0.5382 (0.5382)  acc1: 87.5000 (87.5000)  acc5: 96.8750 (96.8750)  time: 1.0367  data: 0.8757  max mem: 17867
[18:29:31.853371] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.4541 (1.4566)  acc1: 59.3750 (62.6684)  acc5: 87.5000 (88.5292)  time: 0.1299  data: 0.0002  max mem: 17867
[18:30:36.825343] Test:  [1000/1563]  eta: 0:01:13  loss: 2.2820 (1.6829)  acc1: 43.7500 (59.1783)  acc5: 75.0000 (84.5873)  time: 0.1299  data: 0.0002  max mem: 17867
[18:31:41.824099] Test:  [1500/1563]  eta: 0:00:08  loss: 1.0621 (1.8105)  acc1: 75.0000 (57.1994)  acc5: 90.6250 (82.3805)  time: 0.1299  data: 0.0002  max mem: 17867
[18:31:49.806964] Test:  [1562/1563]  eta: 0:00:00  loss: 0.7797 (1.8057)  acc1: 81.2500 (57.3320)  acc5: 93.7500 (82.4900)  time: 0.1262  data: 0.0001  max mem: 17867
[18:31:49.887940] Test: Total time: 0:03:24 (0.1306 s / it)
[18:31:50.246111] * Acc@1 57.332 Acc@5 82.490 loss 1.806
[18:31:50.246288] Accuracy of the network on the 50000 test images: 57.3%
[18:31:50.246311] Max accuracy: 57.33%
[18:31:50.321325] log_dir: ./output_dir_cml_spikformer
[18:31:51.672736] Epoch: [14]  [   0/5004]  eta: 1:52:37  lr: 0.001194  loss: 3.3897 (3.3897)  time: 1.3504  data: 0.8933  max mem: 17867
[18:43:26.567303] Epoch: [14]  [2000/5004]  eta: 0:17:25  lr: 0.001193  loss: 3.4907 (3.6282)  time: 0.3476  data: 0.0002  max mem: 17867
[18:55:00.442398] Epoch: [14]  [4000/5004]  eta: 0:05:48  lr: 0.001193  loss: 3.7483 (3.6312)  time: 0.3478  data: 0.0002  max mem: 17867
[19:00:47.740415] Epoch: [14]  [5003/5004]  eta: 0:00:00  lr: 0.001192  loss: 3.6445 (3.6303)  time: 0.3408  data: 0.0011  max mem: 17867
[19:00:48.063730] Epoch: [14] Total time: 0:28:57 (0.3473 s / it)
[19:00:48.065868] Averaged stats: lr: 0.001192  loss: 3.6445 (3.6300)
[19:00:49.107621] Test:  [   0/1563]  eta: 0:27:00  loss: 0.7595 (0.7595)  acc1: 84.3750 (84.3750)  acc5: 93.7500 (93.7500)  time: 1.0369  data: 0.9012  max mem: 17867
[19:01:54.412765] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.6146 (1.4701)  acc1: 53.1250 (62.7308)  acc5: 87.5000 (88.2984)  time: 0.1296  data: 0.0002  max mem: 17867
[19:02:59.555181] Test:  [1000/1563]  eta: 0:01:13  loss: 1.8499 (1.6596)  acc1: 53.1250 (59.8932)  acc5: 87.5000 (84.6934)  time: 0.1299  data: 0.0002  max mem: 17867
[19:04:04.548013] Test:  [1500/1563]  eta: 0:00:08  loss: 1.2196 (1.7890)  acc1: 65.6250 (57.5700)  acc5: 90.6250 (82.6074)  time: 0.1300  data: 0.0002  max mem: 17867
[19:04:12.530507] Test:  [1562/1563]  eta: 0:00:00  loss: 0.7928 (1.7863)  acc1: 84.3750 (57.7040)  acc5: 93.7500 (82.6440)  time: 0.1262  data: 0.0001  max mem: 17867
[19:04:12.597993] Test: Total time: 0:03:24 (0.1309 s / it)
[19:04:13.061346] * Acc@1 57.704 Acc@5 82.644 loss 1.786
[19:04:13.061489] Accuracy of the network on the 50000 test images: 57.7%
[19:04:13.061510] Max accuracy: 57.70%
[19:04:13.085815] log_dir: ./output_dir_cml_spikformer
[19:04:14.505270] Epoch: [15]  [   0/5004]  eta: 1:58:19  lr: 0.001192  loss: 2.9277 (2.9277)  time: 1.4187  data: 0.9189  max mem: 17867
[19:15:49.860251] Epoch: [15]  [2000/5004]  eta: 0:17:25  lr: 0.001192  loss: 3.5422 (3.6184)  time: 0.3519  data: 0.0002  max mem: 17867
[19:27:25.224693] Epoch: [15]  [4000/5004]  eta: 0:05:49  lr: 0.001191  loss: 3.7006 (3.6145)  time: 0.3483  data: 0.0002  max mem: 17867
[19:33:13.498566] Epoch: [15]  [5003/5004]  eta: 0:00:00  lr: 0.001191  loss: 3.2938 (3.6139)  time: 0.3440  data: 0.0011  max mem: 17867
[19:33:13.822313] Epoch: [15] Total time: 0:29:00 (0.3479 s / it)
[19:33:13.823182] Averaged stats: lr: 0.001191  loss: 3.2938 (3.6018)
[19:33:14.858879] Test:  [   0/1563]  eta: 0:26:51  loss: 0.5482 (0.5482)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.0310  data: 0.8854  max mem: 17867
[19:34:20.050878] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.5121 (1.4071)  acc1: 56.2500 (64.1280)  acc5: 90.6250 (89.2216)  time: 0.1299  data: 0.0002  max mem: 17867
[19:35:25.033046] Test:  [1000/1563]  eta: 0:01:13  loss: 1.5125 (1.6195)  acc1: 59.3750 (60.5613)  acc5: 87.5000 (85.5176)  time: 0.1299  data: 0.0002  max mem: 17867
[19:36:30.077105] Test:  [1500/1563]  eta: 0:00:08  loss: 1.1868 (1.7539)  acc1: 71.8750 (58.0988)  acc5: 90.6250 (83.1925)  time: 0.1299  data: 0.0002  max mem: 17867
[19:36:38.059418] Test:  [1562/1563]  eta: 0:00:00  loss: 0.8815 (1.7518)  acc1: 78.1250 (58.1800)  acc5: 93.7500 (83.2380)  time: 0.1262  data: 0.0001  max mem: 17867
[19:36:38.124238] Test: Total time: 0:03:24 (0.1307 s / it)
[19:36:38.274094] * Acc@1 58.180 Acc@5 83.238 loss 1.752
[19:36:38.274238] Accuracy of the network on the 50000 test images: 58.2%
[19:36:38.274260] Max accuracy: 58.18%
[19:36:38.313368] log_dir: ./output_dir_cml_spikformer
[19:36:39.963473] Epoch: [16]  [   0/5004]  eta: 2:17:32  lr: 0.001191  loss: 2.8556 (2.8556)  time: 1.6491  data: 0.9715  max mem: 17867
[19:48:15.328109] Epoch: [16]  [2000/5004]  eta: 0:17:26  lr: 0.001190  loss: 3.5451 (3.5941)  time: 0.3446  data: 0.0002  max mem: 17867
[19:59:51.046915] Epoch: [16]  [4000/5004]  eta: 0:05:49  lr: 0.001189  loss: 3.3264 (3.5837)  time: 0.3520  data: 0.0002  max mem: 17867
[20:05:40.281760] Epoch: [16]  [5003/5004]  eta: 0:00:00  lr: 0.001189  loss: 3.5179 (3.5816)  time: 0.3429  data: 0.0006  max mem: 17867
[20:05:40.740161] Epoch: [16] Total time: 0:29:02 (0.3482 s / it)
[20:05:40.745191] Averaged stats: lr: 0.001189  loss: 3.5179 (3.5767)
[20:05:42.620684] Test:  [   0/1563]  eta: 0:48:45  loss: 0.6413 (0.6413)  acc1: 90.6250 (90.6250)  acc5: 93.7500 (93.7500)  time: 1.8717  data: 1.7322  max mem: 17867
[20:06:47.725455] Test:  [ 500/1563]  eta: 0:02:22  loss: 1.8012 (1.5711)  acc1: 59.3750 (61.7141)  acc5: 84.3750 (86.7889)  time: 0.1299  data: 0.0002  max mem: 17867
[20:07:52.706466] Test:  [1000/1563]  eta: 0:01:14  loss: 1.8418 (1.7349)  acc1: 50.0000 (58.8380)  acc5: 84.3750 (83.9348)  time: 0.1299  data: 0.0002  max mem: 17867
[20:08:57.686615] Test:  [1500/1563]  eta: 0:00:08  loss: 1.3312 (1.8382)  acc1: 71.8750 (56.9350)  acc5: 87.5000 (82.1015)  time: 0.1299  data: 0.0002  max mem: 17867
[20:09:05.674485] Test:  [1562/1563]  eta: 0:00:00  loss: 1.3487 (1.8388)  acc1: 71.8750 (56.9740)  acc5: 87.5000 (82.0900)  time: 0.1262  data: 0.0001  max mem: 17867
[20:09:05.739328] Test: Total time: 0:03:24 (0.1312 s / it)
[20:09:05.982794] * Acc@1 56.974 Acc@5 82.090 loss 1.839
[20:09:05.983000] Accuracy of the network on the 50000 test images: 57.0%
[20:09:05.983028] Max accuracy: 58.18%
[20:09:06.009950] log_dir: ./output_dir_cml_spikformer
[20:09:07.441044] Epoch: [17]  [   0/5004]  eta: 1:59:16  lr: 0.001189  loss: 3.9312 (3.9312)  time: 1.4302  data: 1.0165  max mem: 17867
[20:20:42.709640] Epoch: [17]  [2000/5004]  eta: 0:17:25  lr: 0.001188  loss: 3.6026 (3.5572)  time: 0.3515  data: 0.0002  max mem: 17867
[20:32:21.061972] Epoch: [17]  [4000/5004]  eta: 0:05:50  lr: 0.001187  loss: 3.4599 (3.5538)  time: 0.3439  data: 0.0002  max mem: 17867
[20:38:09.641070] Epoch: [17]  [5003/5004]  eta: 0:00:00  lr: 0.001187  loss: 3.4938 (3.5565)  time: 0.3425  data: 0.0011  max mem: 17867
[20:38:09.961318] Epoch: [17] Total time: 0:29:03 (0.3485 s / it)
[20:38:09.962039] Averaged stats: lr: 0.001187  loss: 3.4938 (3.5550)
[20:38:10.897318] Test:  [   0/1563]  eta: 0:24:16  loss: 0.4278 (0.4278)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 0.9316  data: 0.7930  max mem: 17867
[20:39:15.931994] Test:  [ 500/1563]  eta: 0:02:19  loss: 1.6781 (1.4506)  acc1: 62.5000 (63.4731)  acc5: 84.3750 (88.9970)  time: 0.1297  data: 0.0002  max mem: 17867
[20:40:20.854456] Test:  [1000/1563]  eta: 0:01:13  loss: 2.0310 (1.6583)  acc1: 50.0000 (60.3990)  acc5: 81.2500 (85.3740)  time: 0.1297  data: 0.0002  max mem: 17867
[20:41:25.773100] Test:  [1500/1563]  eta: 0:00:08  loss: 1.2593 (1.7870)  acc1: 71.8750 (57.9988)  acc5: 87.5000 (83.2237)  time: 0.1298  data: 0.0002  max mem: 17867
[20:41:33.748815] Test:  [1562/1563]  eta: 0:00:00  loss: 0.7654 (1.7783)  acc1: 84.3750 (58.1900)  acc5: 93.7500 (83.3660)  time: 0.1261  data: 0.0001  max mem: 17867
[20:41:33.817251] Test: Total time: 0:03:23 (0.1304 s / it)
[20:41:34.222255] * Acc@1 58.190 Acc@5 83.366 loss 1.778
[20:41:34.222399] Accuracy of the network on the 50000 test images: 58.2%
[20:41:34.222420] Max accuracy: 58.19%
[20:41:34.244195] log_dir: ./output_dir_cml_spikformer
[20:41:35.796559] Epoch: [18]  [   0/5004]  eta: 2:09:24  lr: 0.001187  loss: 3.6723 (3.6723)  time: 1.5516  data: 0.8836  max mem: 17867
[20:53:11.765302] Epoch: [18]  [2000/5004]  eta: 0:17:27  lr: 0.001186  loss: 3.4936 (3.5298)  time: 0.3445  data: 0.0002  max mem: 17867
[21:04:46.149628] Epoch: [18]  [4000/5004]  eta: 0:05:49  lr: 0.001185  loss: 3.6290 (3.5253)  time: 0.3464  data: 0.0002  max mem: 17867
[21:10:35.299426] Epoch: [18]  [5003/5004]  eta: 0:00:00  lr: 0.001185  loss: 3.4063 (3.5279)  time: 0.3440  data: 0.0006  max mem: 17867
[21:10:35.616376] Epoch: [18] Total time: 0:29:01 (0.3480 s / it)
[21:10:35.625375] Averaged stats: lr: 0.001185  loss: 3.4063 (3.5361)
[21:10:36.791769] Test:  [   0/1563]  eta: 0:30:17  loss: 0.7091 (0.7091)  acc1: 87.5000 (87.5000)  acc5: 90.6250 (90.6250)  time: 1.1627  data: 0.9925  max mem: 17867
[21:11:41.869008] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.4745 (1.3882)  acc1: 62.5000 (65.0449)  acc5: 90.6250 (89.5709)  time: 0.1304  data: 0.0002  max mem: 17867
[21:12:46.841859] Test:  [1000/1563]  eta: 0:01:13  loss: 2.0140 (1.5767)  acc1: 56.2500 (62.0692)  acc5: 81.2500 (86.1233)  time: 0.1298  data: 0.0002  max mem: 17867
[21:13:51.806519] Test:  [1500/1563]  eta: 0:00:08  loss: 1.2932 (1.7029)  acc1: 71.8750 (59.5832)  acc5: 90.6250 (84.1148)  time: 0.1298  data: 0.0002  max mem: 17867
[21:13:59.785592] Test:  [1562/1563]  eta: 0:00:00  loss: 1.0783 (1.6962)  acc1: 78.1250 (59.7480)  acc5: 93.7500 (84.2180)  time: 0.1262  data: 0.0001  max mem: 17867
[21:13:59.852076] Test: Total time: 0:03:24 (0.1307 s / it)
[21:14:00.320823] * Acc@1 59.748 Acc@5 84.218 loss 1.696
[21:14:00.320971] Accuracy of the network on the 50000 test images: 59.7%
[21:14:00.320993] Max accuracy: 59.75%
[21:14:00.333593] log_dir: ./output_dir_cml_spikformer
[21:14:01.865780] Epoch: [19]  [   0/5004]  eta: 2:07:43  lr: 0.001185  loss: 3.3643 (3.3643)  time: 1.5314  data: 0.9065  max mem: 17867
[21:25:38.749402] Epoch: [19]  [2000/5004]  eta: 0:17:28  lr: 0.001184  loss: 3.6603 (3.5084)  time: 0.3468  data: 0.0002  max mem: 17867
[21:37:14.548149] Epoch: [19]  [4000/5004]  eta: 0:05:49  lr: 0.001183  loss: 3.4784 (3.5242)  time: 0.3463  data: 0.0002  max mem: 17867
[21:43:03.406818] Epoch: [19]  [5003/5004]  eta: 0:00:00  lr: 0.001183  loss: 3.5444 (3.5209)  time: 0.3432  data: 0.0006  max mem: 17867
[21:43:03.751346] Epoch: [19] Total time: 0:29:03 (0.3484 s / it)
[21:43:03.756309] Averaged stats: lr: 0.001183  loss: 3.5444 (3.5174)
[21:43:04.738629] Test:  [   0/1563]  eta: 0:25:29  loss: 0.8898 (0.8898)  acc1: 78.1250 (78.1250)  acc5: 93.7500 (93.7500)  time: 0.9789  data: 0.8347  max mem: 17867
[21:44:09.742442] Test:  [ 500/1563]  eta: 0:02:19  loss: 1.2649 (1.4660)  acc1: 59.3750 (63.8535)  acc5: 93.7500 (88.8161)  time: 0.1299  data: 0.0002  max mem: 17867
[21:45:14.768551] Test:  [1000/1563]  eta: 0:01:13  loss: 2.1931 (1.6554)  acc1: 40.6250 (60.8173)  acc5: 84.3750 (85.3834)  time: 0.1300  data: 0.0002  max mem: 17867
[21:46:19.749191] Test:  [1500/1563]  eta: 0:00:08  loss: 1.1332 (1.7633)  acc1: 75.0000 (58.8941)  acc5: 87.5000 (83.4340)  time: 0.1299  data: 0.0002  max mem: 17867
[21:46:27.834077] Test:  [1562/1563]  eta: 0:00:00  loss: 1.1536 (1.7634)  acc1: 75.0000 (58.9100)  acc5: 90.6250 (83.4760)  time: 0.1313  data: 0.0001  max mem: 17867
[21:46:27.924071] Test: Total time: 0:03:24 (0.1306 s / it)
[21:46:28.013329] * Acc@1 58.910 Acc@5 83.476 loss 1.763
[21:46:28.013531] Accuracy of the network on the 50000 test images: 58.9%
[21:46:28.013553] Max accuracy: 59.75%
[21:46:28.050262] log_dir: ./output_dir_cml_spikformer
[21:46:29.522995] Epoch: [20]  [   0/5004]  eta: 2:02:45  lr: 0.001183  loss: 3.4407 (3.4407)  time: 1.4719  data: 1.0862  max mem: 17867
[21:58:07.018212] Epoch: [20]  [2000/5004]  eta: 0:17:29  lr: 0.001182  loss: 3.4145 (3.4954)  time: 0.3472  data: 0.0002  max mem: 17867
[22:09:43.023307] Epoch: [20]  [4000/5004]  eta: 0:05:50  lr: 0.001181  loss: 3.4272 (3.4956)  time: 0.3469  data: 0.0002  max mem: 17867
[22:15:31.951036] Epoch: [20]  [5003/5004]  eta: 0:00:00  lr: 0.001180  loss: 3.4951 (3.5013)  time: 0.3438  data: 0.0006  max mem: 17867
[22:15:32.299125] Epoch: [20] Total time: 0:29:04 (0.3486 s / it)
[22:15:32.299799] Averaged stats: lr: 0.001180  loss: 3.4951 (3.4992)
[22:15:33.226423] Test:  [   0/1563]  eta: 0:24:02  loss: 0.5980 (0.5980)  acc1: 90.6250 (90.6250)  acc5: 96.8750 (96.8750)  time: 0.9229  data: 0.7856  max mem: 17867
[22:16:38.375010] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.4259 (1.3458)  acc1: 56.2500 (65.5501)  acc5: 87.5000 (89.4336)  time: 0.1299  data: 0.0002  max mem: 17867
[22:17:43.345740] Test:  [1000/1563]  eta: 0:01:13  loss: 1.7602 (1.5482)  acc1: 56.2500 (62.0067)  acc5: 81.2500 (86.0109)  time: 0.1300  data: 0.0002  max mem: 17867
[22:18:48.349181] Test:  [1500/1563]  eta: 0:00:08  loss: 0.9806 (1.6650)  acc1: 75.0000 (59.9017)  acc5: 93.7500 (84.0044)  time: 0.1299  data: 0.0002  max mem: 17867
[22:18:56.331964] Test:  [1562/1563]  eta: 0:00:00  loss: 0.7747 (1.6614)  acc1: 84.3750 (59.9980)  acc5: 96.8750 (84.0820)  time: 0.1262  data: 0.0001  max mem: 17867
[22:18:56.415124] Test: Total time: 0:03:24 (0.1306 s / it)
[22:18:56.510068] * Acc@1 59.998 Acc@5 84.082 loss 1.661
[22:18:56.510203] Accuracy of the network on the 50000 test images: 60.0%
[22:18:56.510223] Max accuracy: 60.00%
[22:18:56.535428] log_dir: ./output_dir_cml_spikformer
[22:18:58.312373] Epoch: [21]  [   0/5004]  eta: 2:28:01  lr: 0.001180  loss: 3.5381 (3.5381)  time: 1.7749  data: 1.0715  max mem: 17867
[22:30:34.873903] Epoch: [21]  [2000/5004]  eta: 0:17:28  lr: 0.001179  loss: 3.5853 (3.4821)  time: 0.3436  data: 0.0001  max mem: 17867
[22:42:12.448908] Epoch: [21]  [4000/5004]  eta: 0:05:50  lr: 0.001178  loss: 3.5236 (3.4828)  time: 0.3513  data: 0.0002  max mem: 17867
[22:48:01.421847] Epoch: [21]  [5003/5004]  eta: 0:00:00  lr: 0.001178  loss: 3.4076 (3.4857)  time: 0.3461  data: 0.0006  max mem: 17867
[22:48:01.775724] Epoch: [21] Total time: 0:29:05 (0.3488 s / it)
[22:48:01.779554] Averaged stats: lr: 0.001178  loss: 3.4076 (3.4864)
[22:48:02.840695] Test:  [   0/1563]  eta: 0:27:33  loss: 0.8343 (0.8343)  acc1: 84.3750 (84.3750)  acc5: 96.8750 (96.8750)  time: 1.0577  data: 0.9203  max mem: 17867
[22:49:07.923968] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.2922 (1.3164)  acc1: 65.6250 (66.1739)  acc5: 87.5000 (90.2133)  time: 0.1299  data: 0.0002  max mem: 17867
[22:50:12.885231] Test:  [1000/1563]  eta: 0:01:13  loss: 1.8744 (1.5247)  acc1: 46.8750 (62.8777)  acc5: 84.3750 (86.6415)  time: 0.1298  data: 0.0002  max mem: 17867
[22:51:17.848952] Test:  [1500/1563]  eta: 0:00:08  loss: 1.0931 (1.6606)  acc1: 71.8750 (60.3202)  acc5: 90.6250 (84.3396)  time: 0.1299  data: 0.0002  max mem: 17867
[22:51:25.832346] Test:  [1562/1563]  eta: 0:00:00  loss: 0.9341 (1.6563)  acc1: 78.1250 (60.3120)  acc5: 93.7500 (84.3880)  time: 0.1263  data: 0.0001  max mem: 17867
[22:51:25.892337] Test: Total time: 0:03:24 (0.1306 s / it)
[22:51:26.141014] * Acc@1 60.312 Acc@5 84.388 loss 1.656
[22:51:26.141156] Accuracy of the network on the 50000 test images: 60.3%
[22:51:26.141176] Max accuracy: 60.31%
[22:51:26.167184] log_dir: ./output_dir_cml_spikformer
[22:51:27.618792] Epoch: [22]  [   0/5004]  eta: 2:00:56  lr: 0.001178  loss: 3.1713 (3.1713)  time: 1.4502  data: 1.0431  max mem: 17867
[23:03:04.523639] Epoch: [22]  [2000/5004]  eta: 0:17:28  lr: 0.001177  loss: 3.5858 (3.4731)  time: 0.3475  data: 0.0002  max mem: 17867
[23:14:41.789685] Epoch: [22]  [4000/5004]  eta: 0:05:50  lr: 0.001176  loss: 3.6473 (3.4780)  time: 0.3495  data: 0.0002  max mem: 17867
[23:20:31.330726] Epoch: [22]  [5003/5004]  eta: 0:00:00  lr: 0.001175  loss: 3.5107 (3.4762)  time: 0.3479  data: 0.0011  max mem: 17867
[23:20:31.669453] Epoch: [22] Total time: 0:29:05 (0.3488 s / it)
[23:20:31.670459] Averaged stats: lr: 0.001175  loss: 3.5107 (3.4720)
[23:20:32.837474] Test:  [   0/1563]  eta: 0:30:17  loss: 0.4232 (0.4232)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.1631  data: 1.0256  max mem: 17867
[23:21:37.832646] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.0983 (1.3423)  acc1: 65.6250 (66.1302)  acc5: 90.6250 (89.9763)  time: 0.1299  data: 0.0002  max mem: 17867
[23:22:42.798747] Test:  [1000/1563]  eta: 0:01:13  loss: 1.7511 (1.5202)  acc1: 56.2500 (63.1306)  acc5: 84.3750 (86.8881)  time: 0.1299  data: 0.0002  max mem: 17867
[23:23:47.750390] Test:  [1500/1563]  eta: 0:00:08  loss: 0.9276 (1.6342)  acc1: 78.1250 (60.8990)  acc5: 93.7500 (85.1141)  time: 0.1299  data: 0.0002  max mem: 17867
[23:23:55.729854] Test:  [1562/1563]  eta: 0:00:00  loss: 0.8174 (1.6324)  acc1: 81.2500 (60.9400)  acc5: 93.7500 (85.1520)  time: 0.1262  data: 0.0001  max mem: 17867
[23:23:55.789954] Test: Total time: 0:03:24 (0.1306 s / it)
[23:23:56.228323] * Acc@1 60.940 Acc@5 85.152 loss 1.632
[23:23:56.228464] Accuracy of the network on the 50000 test images: 60.9%
[23:23:56.228484] Max accuracy: 60.94%
[23:23:56.245302] log_dir: ./output_dir_cml_spikformer
[23:23:57.737166] Epoch: [23]  [   0/5004]  eta: 2:04:21  lr: 0.001175  loss: 3.3923 (3.3923)  time: 1.4912  data: 1.0514  max mem: 17867
[23:35:35.616600] Epoch: [23]  [2000/5004]  eta: 0:17:29  lr: 0.001174  loss: 3.3436 (3.4576)  time: 0.3488  data: 0.0002  max mem: 17867
[23:47:12.692853] Epoch: [23]  [4000/5004]  eta: 0:05:50  lr: 0.001173  loss: 3.4858 (3.4575)  time: 0.3478  data: 0.0002  max mem: 17867
[23:53:02.600743] Epoch: [23]  [5003/5004]  eta: 0:00:00  lr: 0.001172  loss: 3.4519 (3.4589)  time: 0.3440  data: 0.0006  max mem: 17867
[23:53:02.946067] Epoch: [23] Total time: 0:29:06 (0.3491 s / it)
[23:53:02.951361] Averaged stats: lr: 0.001172  loss: 3.4519 (3.4574)
[23:53:03.951701] Test:  [   0/1563]  eta: 0:25:58  loss: 1.1722 (1.1722)  acc1: 81.2500 (81.2500)  acc5: 90.6250 (90.6250)  time: 0.9969  data: 0.8526  max mem: 17867
[23:54:08.971706] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.6599 (1.3777)  acc1: 59.3750 (65.3568)  acc5: 90.6250 (89.9701)  time: 0.1299  data: 0.0002  max mem: 17867
[23:55:13.968316] Test:  [1000/1563]  eta: 0:01:13  loss: 1.5005 (1.5459)  acc1: 56.2500 (62.7466)  acc5: 90.6250 (86.7695)  time: 0.1299  data: 0.0002  max mem: 17867
[23:56:18.964488] Test:  [1500/1563]  eta: 0:00:08  loss: 0.9893 (1.6511)  acc1: 78.1250 (60.8136)  acc5: 90.6250 (84.9309)  time: 0.1300  data: 0.0002  max mem: 17867
[23:56:26.947673] Test:  [1562/1563]  eta: 0:00:00  loss: 0.8675 (1.6522)  acc1: 78.1250 (60.8400)  acc5: 90.6250 (84.9060)  time: 0.1263  data: 0.0001  max mem: 17867
[23:56:27.192560] Test: Total time: 0:03:24 (0.1307 s / it)
[23:56:27.209510] * Acc@1 60.840 Acc@5 84.906 loss 1.652
[23:56:27.209643] Accuracy of the network on the 50000 test images: 60.8%
[23:56:27.209667] Max accuracy: 60.94%
[23:56:27.237692] log_dir: ./output_dir_cml_spikformer
[23:56:28.892830] Epoch: [24]  [   0/5004]  eta: 2:17:59  lr: 0.001172  loss: 2.7536 (2.7536)  time: 1.6545  data: 1.1866  max mem: 17867
[00:08:09.437800] Epoch: [24]  [2000/5004]  eta: 0:17:34  lr: 0.001171  loss: 3.3747 (3.4509)  time: 0.3511  data: 0.0002  max mem: 17867
[00:19:49.200305] Epoch: [24]  [4000/5004]  eta: 0:05:51  lr: 0.001170  loss: 3.3663 (3.4446)  time: 0.3543  data: 0.0002  max mem: 17867
[00:25:40.334497] Epoch: [24]  [5003/5004]  eta: 0:00:00  lr: 0.001169  loss: 3.3860 (3.4408)  time: 0.3433  data: 0.0008  max mem: 17867
[00:25:40.657674] Epoch: [24] Total time: 0:29:13 (0.3504 s / it)
[00:25:40.658313] Averaged stats: lr: 0.001169  loss: 3.3860 (3.4427)
[00:25:42.132850] Test:  [   0/1563]  eta: 0:38:18  loss: 0.5553 (0.5553)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.4708  data: 1.3336  max mem: 17867
[00:26:47.212115] Test:  [ 500/1563]  eta: 0:02:21  loss: 1.2652 (1.3687)  acc1: 62.5000 (66.1365)  acc5: 90.6250 (90.5127)  time: 0.1302  data: 0.0002  max mem: 17867
[00:27:52.270739] Test:  [1000/1563]  eta: 0:01:14  loss: 1.8216 (1.5335)  acc1: 46.8750 (63.5302)  acc5: 87.5000 (87.3876)  time: 0.1302  data: 0.0005  max mem: 17867
[00:28:57.320737] Test:  [1500/1563]  eta: 0:00:08  loss: 0.9392 (1.6444)  acc1: 78.1250 (61.4819)  acc5: 90.6250 (85.4514)  time: 0.1298  data: 0.0002  max mem: 17867
[00:29:05.301099] Test:  [1562/1563]  eta: 0:00:00  loss: 0.9138 (1.6406)  acc1: 81.2500 (61.6420)  acc5: 93.7500 (85.4940)  time: 0.1261  data: 0.0001  max mem: 17867
[00:29:05.384930] Test: Total time: 0:03:24 (0.1310 s / it)
[00:29:05.411234] * Acc@1 61.642 Acc@5 85.494 loss 1.641
[00:29:05.411393] Accuracy of the network on the 50000 test images: 61.6%
[00:29:05.411423] Max accuracy: 61.64%
[00:29:05.452020] log_dir: ./output_dir_cml_spikformer
[00:29:06.954210] Epoch: [25]  [   0/5004]  eta: 2:05:09  lr: 0.001169  loss: 3.0243 (3.0243)  time: 1.5006  data: 0.9808  max mem: 17867
[00:40:46.716331] Epoch: [25]  [2000/5004]  eta: 0:17:32  lr: 0.001168  loss: 3.3247 (3.4427)  time: 0.3505  data: 0.0002  max mem: 17867
[00:52:25.718962] Epoch: [25]  [4000/5004]  eta: 0:05:51  lr: 0.001167  loss: 3.2914 (3.4368)  time: 0.3475  data: 0.0002  max mem: 17867
[00:58:16.502449] Epoch: [25]  [5003/5004]  eta: 0:00:00  lr: 0.001166  loss: 3.3511 (3.4360)  time: 0.3513  data: 0.0007  max mem: 17867
[00:58:16.838020] Epoch: [25] Total time: 0:29:11 (0.3500 s / it)
[00:58:16.849864] Averaged stats: lr: 0.001166  loss: 3.3511 (3.4314)
[00:58:17.777019] Test:  [   0/1563]  eta: 0:24:03  loss: 0.8717 (0.8717)  acc1: 78.1250 (78.1250)  acc5: 96.8750 (96.8750)  time: 0.9237  data: 0.7875  max mem: 17867
[00:59:22.792666] Test:  [ 500/1563]  eta: 0:02:19  loss: 1.6004 (1.3478)  acc1: 59.3750 (66.1427)  acc5: 90.6250 (90.3256)  time: 0.1299  data: 0.0002  max mem: 17867
[01:00:27.789439] Test:  [1000/1563]  eta: 0:01:13  loss: 1.8821 (1.4889)  acc1: 56.2500 (64.1359)  acc5: 87.5000 (87.5343)  time: 0.1300  data: 0.0002  max mem: 17867
[01:01:32.799920] Test:  [1500/1563]  eta: 0:00:08  loss: 1.1693 (1.6065)  acc1: 75.0000 (61.7130)  acc5: 87.5000 (85.6492)  time: 0.1300  data: 0.0003  max mem: 17867
[01:01:40.791298] Test:  [1562/1563]  eta: 0:00:00  loss: 0.7967 (1.6099)  acc1: 81.2500 (61.6740)  acc5: 93.7500 (85.6180)  time: 0.1262  data: 0.0001  max mem: 17867
[01:01:40.855389] Test: Total time: 0:03:24 (0.1305 s / it)
[01:01:41.442176] * Acc@1 61.674 Acc@5 85.618 loss 1.610
[01:01:41.442323] Accuracy of the network on the 50000 test images: 61.7%
[01:01:41.442344] Max accuracy: 61.67%
[01:01:41.469681] log_dir: ./output_dir_cml_spikformer
[01:01:42.947656] Epoch: [26]  [   0/5004]  eta: 2:03:12  lr: 0.001166  loss: 3.2718 (3.2718)  time: 1.4773  data: 0.9720  max mem: 17867
[01:13:23.911755] Epoch: [26]  [2000/5004]  eta: 0:17:34  lr: 0.001165  loss: 3.3696 (3.4091)  time: 0.3481  data: 0.0002  max mem: 17867
[01:25:04.134999] Epoch: [26]  [4000/5004]  eta: 0:05:51  lr: 0.001163  loss: 3.5166 (3.4191)  time: 0.3539  data: 0.0002  max mem: 17867
[01:30:55.103433] Epoch: [26]  [5003/5004]  eta: 0:00:00  lr: 0.001163  loss: 3.3289 (3.4170)  time: 0.3432  data: 0.0011  max mem: 17867
[01:30:55.443403] Epoch: [26] Total time: 0:29:13 (0.3505 s / it)
[01:30:55.460848] Averaged stats: lr: 0.001163  loss: 3.3289 (3.4197)
[01:30:57.298885] Test:  [   0/1563]  eta: 0:47:45  loss: 1.6909 (1.6909)  acc1: 46.8750 (46.8750)  acc5: 87.5000 (87.5000)  time: 1.8334  data: 1.1859  max mem: 17867
[01:32:02.451570] Test:  [ 500/1563]  eta: 0:02:22  loss: 1.2037 (1.2496)  acc1: 62.5000 (68.4256)  acc5: 90.6250 (91.1677)  time: 0.1300  data: 0.0002  max mem: 17867
[01:33:07.443058] Test:  [1000/1563]  eta: 0:01:14  loss: 1.7023 (1.4541)  acc1: 53.1250 (64.6791)  acc5: 84.3750 (87.7716)  time: 0.1298  data: 0.0002  max mem: 17867
[01:34:12.416952] Test:  [1500/1563]  eta: 0:00:08  loss: 0.9510 (1.5565)  acc1: 78.1250 (62.7727)  acc5: 93.7500 (86.2217)  time: 0.1300  data: 0.0002  max mem: 17867
[01:34:20.400368] Test:  [1562/1563]  eta: 0:00:00  loss: 0.8701 (1.5557)  acc1: 81.2500 (62.7780)  acc5: 90.6250 (86.2320)  time: 0.1262  data: 0.0001  max mem: 17867
[01:34:20.474333] Test: Total time: 0:03:25 (0.1312 s / it)
[01:34:20.775047] * Acc@1 62.778 Acc@5 86.232 loss 1.556
[01:34:20.775222] Accuracy of the network on the 50000 test images: 62.8%
[01:34:20.775246] Max accuracy: 62.78%
[01:34:20.819373] log_dir: ./output_dir_cml_spikformer
[01:34:22.274805] Epoch: [27]  [   0/5004]  eta: 2:01:17  lr: 0.001163  loss: 3.7722 (3.7722)  time: 1.4544  data: 0.9453  max mem: 17867
[01:46:02.613174] Epoch: [27]  [2000/5004]  eta: 0:17:33  lr: 0.001161  loss: 3.5383 (3.4001)  time: 0.3553  data: 0.0002  max mem: 17867
[01:57:42.526462] Epoch: [27]  [4000/5004]  eta: 0:05:51  lr: 0.001160  loss: 3.2756 (3.4049)  time: 0.3470  data: 0.0002  max mem: 17867
[02:03:33.227196] Epoch: [27]  [5003/5004]  eta: 0:00:00  lr: 0.001159  loss: 3.3746 (3.4083)  time: 0.3444  data: 0.0011  max mem: 17867
[02:03:33.616126] Epoch: [27] Total time: 0:29:12 (0.3503 s / it)
[02:03:33.616899] Averaged stats: lr: 0.001159  loss: 3.3746 (3.4100)
[02:03:34.621255] Test:  [   0/1563]  eta: 0:26:04  loss: 0.2611 (0.2611)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.0009  data: 0.8634  max mem: 17867
[02:04:39.601950] Test:  [ 500/1563]  eta: 0:02:19  loss: 1.2779 (1.3112)  acc1: 59.3750 (66.8663)  acc5: 90.6250 (90.5813)  time: 0.1299  data: 0.0002  max mem: 17867
[02:05:44.590690] Test:  [1000/1563]  eta: 0:01:13  loss: 2.0156 (1.4565)  acc1: 53.1250 (64.4293)  acc5: 84.3750 (87.9839)  time: 0.1299  data: 0.0002  max mem: 17867
[02:06:49.570100] Test:  [1500/1563]  eta: 0:00:08  loss: 1.0031 (1.5530)  acc1: 78.1250 (62.5416)  acc5: 93.7500 (86.3674)  time: 0.1300  data: 0.0002  max mem: 17867
[02:06:57.550653] Test:  [1562/1563]  eta: 0:00:00  loss: 0.8547 (1.5557)  acc1: 78.1250 (62.4960)  acc5: 93.7500 (86.3360)  time: 0.1262  data: 0.0001  max mem: 17867
[02:06:57.610184] Test: Total time: 0:03:23 (0.1305 s / it)
[02:06:57.926686] * Acc@1 62.496 Acc@5 86.336 loss 1.556
[02:06:57.926842] Accuracy of the network on the 50000 test images: 62.5%
[02:06:57.926863] Max accuracy: 62.78%
[02:06:57.951100] log_dir: ./output_dir_cml_spikformer
[02:06:59.511163] Epoch: [28]  [   0/5004]  eta: 2:10:03  lr: 0.001159  loss: 3.5414 (3.5414)  time: 1.5594  data: 0.9173  max mem: 17867
[02:18:40.939457] Epoch: [28]  [2000/5004]  eta: 0:17:35  lr: 0.001158  loss: 3.6357 (3.3963)  time: 0.3455  data: 0.0002  max mem: 17867
[02:30:21.805744] Epoch: [28]  [4000/5004]  eta: 0:05:52  lr: 0.001156  loss: 3.2748 (3.3912)  time: 0.3577  data: 0.0002  max mem: 17867
[02:36:12.464283] Epoch: [28]  [5003/5004]  eta: 0:00:00  lr: 0.001156  loss: 3.4245 (3.3913)  time: 0.3448  data: 0.0012  max mem: 17867
[02:36:12.813984] Epoch: [28] Total time: 0:29:14 (0.3507 s / it)
[02:36:12.864594] Averaged stats: lr: 0.001156  loss: 3.4245 (3.3977)
[02:36:13.967266] Test:  [   0/1563]  eta: 0:28:37  loss: 0.5970 (0.5970)  acc1: 90.6250 (90.6250)  acc5: 93.7500 (93.7500)  time: 1.0991  data: 0.9056  max mem: 17867
[02:37:18.969391] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.2043 (1.3089)  acc1: 65.6250 (66.3236)  acc5: 93.7500 (90.0823)  time: 0.1299  data: 0.0002  max mem: 17867
[02:38:23.963821] Test:  [1000/1563]  eta: 0:01:13  loss: 1.5318 (1.4813)  acc1: 59.3750 (63.5552)  acc5: 90.6250 (87.1098)  time: 0.1299  data: 0.0002  max mem: 17867
[02:39:28.978687] Test:  [1500/1563]  eta: 0:00:08  loss: 0.9404 (1.5808)  acc1: 81.2500 (61.9753)  acc5: 93.7500 (85.5784)  time: 0.1300  data: 0.0002  max mem: 17867
[02:39:36.970221] Test:  [1562/1563]  eta: 0:00:00  loss: 0.7668 (1.5804)  acc1: 84.3750 (62.0120)  acc5: 96.8750 (85.5840)  time: 0.1263  data: 0.0001  max mem: 17867
[02:39:37.039200] Test: Total time: 0:03:24 (0.1306 s / it)
[02:39:37.281623] * Acc@1 62.012 Acc@5 85.584 loss 1.580
[02:39:37.281763] Accuracy of the network on the 50000 test images: 62.0%
[02:39:37.281785] Max accuracy: 62.78%
[02:39:37.328006] log_dir: ./output_dir_cml_spikformer
[02:39:38.854195] Epoch: [29]  [   0/5004]  eta: 2:07:11  lr: 0.001156  loss: 3.5139 (3.5139)  time: 1.5251  data: 1.1194  max mem: 17867
[02:51:19.590689] Epoch: [29]  [2000/5004]  eta: 0:17:34  lr: 0.001154  loss: 3.2210 (3.3979)  time: 0.3489  data: 0.0002  max mem: 17867
[03:03:00.140629] Epoch: [29]  [4000/5004]  eta: 0:05:52  lr: 0.001153  loss: 3.4651 (3.3952)  time: 0.3521  data: 0.0002  max mem: 17867
[03:08:51.986354] Epoch: [29]  [5003/5004]  eta: 0:00:00  lr: 0.001152  loss: 3.4407 (3.3948)  time: 0.3454  data: 0.0011  max mem: 17867
[03:08:52.342673] Epoch: [29] Total time: 0:29:15 (0.3507 s / it)
[03:08:52.343666] Averaged stats: lr: 0.001152  loss: 3.4407 (3.3920)
[03:08:53.385267] Test:  [   0/1563]  eta: 0:27:01  loss: 0.5664 (0.5664)  acc1: 90.6250 (90.6250)  acc5: 96.8750 (96.8750)  time: 1.0374  data: 0.8773  max mem: 17867
[03:09:58.460581] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.1012 (1.2196)  acc1: 68.7500 (68.8810)  acc5: 93.7500 (92.0721)  time: 0.1299  data: 0.0002  max mem: 17867
[03:11:03.438155] Test:  [1000/1563]  eta: 0:01:13  loss: 1.4738 (1.4184)  acc1: 65.6250 (65.5969)  acc5: 90.6250 (88.6083)  time: 0.1299  data: 0.0002  max mem: 17867
[03:12:08.444412] Test:  [1500/1563]  eta: 0:00:08  loss: 1.0469 (1.5356)  acc1: 78.1250 (63.3140)  acc5: 93.7500 (86.6651)  time: 0.1300  data: 0.0002  max mem: 17867
[03:12:16.428568] Test:  [1562/1563]  eta: 0:00:00  loss: 0.6088 (1.5321)  acc1: 87.5000 (63.4220)  acc5: 96.8750 (86.7120)  time: 0.1262  data: 0.0001  max mem: 17867
[03:12:16.527894] Test: Total time: 0:03:24 (0.1306 s / it)
[03:12:16.683192] * Acc@1 63.422 Acc@5 86.712 loss 1.532
[03:12:16.683334] Accuracy of the network on the 50000 test images: 63.4%
[03:12:16.683358] Max accuracy: 63.42%
[03:12:16.689924] log_dir: ./output_dir_cml_spikformer
[03:12:18.404810] Epoch: [30]  [   0/5004]  eta: 2:22:58  lr: 0.001152  loss: 2.7499 (2.7499)  time: 1.7143  data: 1.2041  max mem: 17867
[03:23:59.207942] Epoch: [30]  [2000/5004]  eta: 0:17:34  lr: 0.001151  loss: 3.4673 (3.3692)  time: 0.3482  data: 0.0003  max mem: 17867
[03:35:39.314270] Epoch: [30]  [4000/5004]  eta: 0:05:51  lr: 0.001149  loss: 3.3615 (3.3768)  time: 0.3499  data: 0.0002  max mem: 17867
[03:41:30.592610] Epoch: [30]  [5003/5004]  eta: 0:00:00  lr: 0.001148  loss: 3.3847 (3.3785)  time: 0.3448  data: 0.0011  max mem: 17867
[03:41:30.979857] Epoch: [30] Total time: 0:29:14 (0.3506 s / it)
[03:41:30.982095] Averaged stats: lr: 0.001148  loss: 3.3847 (3.3809)
[03:41:31.901427] Test:  [   0/1563]  eta: 0:23:51  loss: 0.5075 (0.5075)  acc1: 90.6250 (90.6250)  acc5: 96.8750 (96.8750)  time: 0.9158  data: 0.7791  max mem: 17867
[03:42:36.912761] Test:  [ 500/1563]  eta: 0:02:19  loss: 1.1798 (1.2431)  acc1: 68.7500 (68.3945)  acc5: 90.6250 (91.2363)  time: 0.1299  data: 0.0002  max mem: 17867
[03:43:41.893994] Test:  [1000/1563]  eta: 0:01:13  loss: 1.7737 (1.3986)  acc1: 50.0000 (65.6843)  acc5: 84.3750 (88.7082)  time: 0.1300  data: 0.0002  max mem: 17867
[03:44:46.978950] Test:  [1500/1563]  eta: 0:00:08  loss: 0.9202 (1.4992)  acc1: 81.2500 (63.7950)  acc5: 93.7500 (87.0003)  time: 0.1305  data: 0.0002  max mem: 17867
[03:44:54.990484] Test:  [1562/1563]  eta: 0:00:00  loss: 0.7738 (1.4985)  acc1: 84.3750 (63.8780)  acc5: 93.7500 (87.0080)  time: 0.1271  data: 0.0001  max mem: 17867
[03:44:55.052212] Test: Total time: 0:03:24 (0.1306 s / it)
[03:44:55.391705] * Acc@1 63.878 Acc@5 87.008 loss 1.498
[03:44:55.391850] Accuracy of the network on the 50000 test images: 63.9%
[03:44:55.391872] Max accuracy: 63.88%
[03:44:55.440996] log_dir: ./output_dir_cml_spikformer
[03:44:57.027705] Epoch: [31]  [   0/5004]  eta: 2:12:12  lr: 0.001148  loss: 3.1932 (3.1932)  time: 1.5853  data: 1.0250  max mem: 17867
[03:56:38.597924] Epoch: [31]  [2000/5004]  eta: 0:17:35  lr: 0.001147  loss: 3.4033 (3.3855)  time: 0.3553  data: 0.0002  max mem: 17867
[04:08:19.284683] Epoch: [31]  [4000/5004]  eta: 0:05:52  lr: 0.001145  loss: 3.2451 (3.3857)  time: 0.3488  data: 0.0002  max mem: 17867
[04:14:10.412360] Epoch: [31]  [5003/5004]  eta: 0:00:00  lr: 0.001144  loss: 3.2701 (3.3827)  time: 0.3451  data: 0.0006  max mem: 17867
[04:14:10.738152] Epoch: [31] Total time: 0:29:15 (0.3508 s / it)
[04:14:10.745296] Averaged stats: lr: 0.001144  loss: 3.2701 (3.3724)
[04:14:11.739356] Test:  [   0/1563]  eta: 0:25:48  loss: 0.9814 (0.9814)  acc1: 84.3750 (84.3750)  acc5: 90.6250 (90.6250)  time: 0.9906  data: 0.8531  max mem: 17867
[04:15:16.864418] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.3213 (1.2688)  acc1: 62.5000 (68.1824)  acc5: 87.5000 (90.9306)  time: 0.1299  data: 0.0002  max mem: 17867
[04:16:21.818536] Test:  [1000/1563]  eta: 0:01:13  loss: 1.5824 (1.4221)  acc1: 62.5000 (65.2504)  acc5: 84.3750 (88.3429)  time: 0.1299  data: 0.0002  max mem: 17867
[04:17:26.786502] Test:  [1500/1563]  eta: 0:00:08  loss: 0.9830 (1.5522)  acc1: 81.2500 (62.7519)  acc5: 90.6250 (86.1988)  time: 0.1299  data: 0.0002  max mem: 17867
[04:17:34.770515] Test:  [1562/1563]  eta: 0:00:00  loss: 0.8114 (1.5464)  acc1: 84.3750 (62.9040)  acc5: 93.7500 (86.2640)  time: 0.1262  data: 0.0001  max mem: 17867
[04:17:34.846606] Test: Total time: 0:03:24 (0.1306 s / it)
[04:17:35.309546] * Acc@1 62.904 Acc@5 86.264 loss 1.546
[04:17:35.309695] Accuracy of the network on the 50000 test images: 62.9%
[04:17:35.309716] Max accuracy: 63.88%
[04:17:35.321981] log_dir: ./output_dir_cml_spikformer
[04:17:36.776464] Epoch: [32]  [   0/5004]  eta: 2:01:14  lr: 0.001144  loss: 3.4219 (3.4219)  time: 1.4537  data: 0.9940  max mem: 17867
[04:29:16.218278] Epoch: [32]  [2000/5004]  eta: 0:17:32  lr: 0.001143  loss: 3.4804 (3.3582)  time: 0.3447  data: 0.0002  max mem: 17867
[04:40:55.031205] Epoch: [32]  [4000/5004]  eta: 0:05:51  lr: 0.001141  loss: 3.3855 (3.3688)  time: 0.3443  data: 0.0002  max mem: 17867
[04:46:45.474247] Epoch: [32]  [5003/5004]  eta: 0:00:00  lr: 0.001140  loss: 3.3690 (3.3713)  time: 0.3451  data: 0.0011  max mem: 17867
[04:46:45.847296] Epoch: [32] Total time: 0:29:10 (0.3498 s / it)
[04:46:45.856497] Averaged stats: lr: 0.001140  loss: 3.3690 (3.3636)
[04:46:46.880299] Test:  [   0/1563]  eta: 0:26:34  loss: 0.3874 (0.3874)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.0203  data: 0.8535  max mem: 17867
[04:47:51.888298] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.3659 (1.2200)  acc1: 65.6250 (69.3800)  acc5: 87.5000 (91.7166)  time: 0.1302  data: 0.0002  max mem: 17867
[04:48:56.856018] Test:  [1000/1563]  eta: 0:01:13  loss: 2.1401 (1.4004)  acc1: 46.8750 (66.2931)  acc5: 84.3750 (88.7925)  time: 0.1301  data: 0.0002  max mem: 17867
[04:50:01.821721] Test:  [1500/1563]  eta: 0:00:08  loss: 1.0907 (1.4959)  acc1: 71.8750 (64.3300)  acc5: 90.6250 (87.2856)  time: 0.1299  data: 0.0002  max mem: 17867
[04:50:09.808654] Test:  [1562/1563]  eta: 0:00:00  loss: 0.7638 (1.4929)  acc1: 81.2500 (64.4000)  acc5: 96.8750 (87.3240)  time: 0.1262  data: 0.0001  max mem: 17867
[04:50:09.872091] Test: Total time: 0:03:24 (0.1305 s / it)
[04:50:10.205794] * Acc@1 64.400 Acc@5 87.324 loss 1.493
[04:50:10.205966] Accuracy of the network on the 50000 test images: 64.4%
[04:50:10.205988] Max accuracy: 64.40%
[04:50:10.221180] log_dir: ./output_dir_cml_spikformer
[04:50:11.678255] Epoch: [33]  [   0/5004]  eta: 2:01:27  lr: 0.001140  loss: 3.9162 (3.9162)  time: 1.4563  data: 0.9329  max mem: 17867
[05:01:52.203795] Epoch: [33]  [2000/5004]  eta: 0:17:33  lr: 0.001138  loss: 3.3037 (3.3406)  time: 0.3541  data: 0.0002  max mem: 17867
[05:13:32.207724] Epoch: [33]  [4000/5004]  eta: 0:05:51  lr: 0.001137  loss: 3.3298 (3.3464)  time: 0.3514  data: 0.0002  max mem: 17867
[05:19:22.990423] Epoch: [33]  [5003/5004]  eta: 0:00:00  lr: 0.001136  loss: 3.3060 (3.3502)  time: 0.3471  data: 0.0011  max mem: 17867
[05:19:23.350062] Epoch: [33] Total time: 0:29:13 (0.3503 s / it)
[05:19:23.350789] Averaged stats: lr: 0.001136  loss: 3.3060 (3.3537)
[05:19:24.473721] Test:  [   0/1563]  eta: 0:29:09  loss: 0.6931 (0.6931)  acc1: 87.5000 (87.5000)  acc5: 96.8750 (96.8750)  time: 1.1195  data: 0.9654  max mem: 17867
[05:20:29.595526] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.1930 (1.2564)  acc1: 59.3750 (68.6128)  acc5: 93.7500 (91.3049)  time: 0.1299  data: 0.0002  max mem: 17867
[05:21:34.583180] Test:  [1000/1563]  eta: 0:01:13  loss: 1.5486 (1.4199)  acc1: 56.2500 (65.7280)  acc5: 84.3750 (88.4397)  time: 0.1299  data: 0.0002  max mem: 17867
[05:22:39.629873] Test:  [1500/1563]  eta: 0:00:08  loss: 0.8414 (1.5122)  acc1: 81.2500 (63.9678)  acc5: 93.7500 (87.0212)  time: 0.1299  data: 0.0002  max mem: 17867
[05:22:47.632633] Test:  [1562/1563]  eta: 0:00:00  loss: 0.8612 (1.5134)  acc1: 81.2500 (63.9660)  acc5: 96.8750 (87.0400)  time: 0.1262  data: 0.0001  max mem: 17867
[05:22:47.722092] Test: Total time: 0:03:24 (0.1308 s / it)
[05:22:47.838552] * Acc@1 63.966 Acc@5 87.040 loss 1.513
[05:22:47.838693] Accuracy of the network on the 50000 test images: 64.0%
[05:22:47.838714] Max accuracy: 64.40%
[05:22:47.845746] log_dir: ./output_dir_cml_spikformer
[05:22:49.316470] Epoch: [34]  [   0/5004]  eta: 2:02:33  lr: 0.001136  loss: 4.1819 (4.1819)  time: 1.4696  data: 0.9906  max mem: 17867
[05:34:29.673510] Epoch: [34]  [2000/5004]  eta: 0:17:33  lr: 0.001134  loss: 3.3223 (3.3425)  time: 0.3503  data: 0.0002  max mem: 17867
[05:46:09.119032] Epoch: [34]  [4000/5004]  eta: 0:05:51  lr: 0.001132  loss: 3.1705 (3.3405)  time: 0.3483  data: 0.0002  max mem: 17867
[05:51:59.919081] Epoch: [34]  [5003/5004]  eta: 0:00:00  lr: 0.001131  loss: 3.1856 (3.3378)  time: 0.3471  data: 0.0011  max mem: 17867
[05:52:00.262104] Epoch: [34] Total time: 0:29:12 (0.3502 s / it)
[05:52:00.266135] Averaged stats: lr: 0.001131  loss: 3.1856 (3.3447)
[05:52:01.223514] Test:  [   0/1563]  eta: 0:24:48  loss: 1.1519 (1.1519)  acc1: 78.1250 (78.1250)  acc5: 93.7500 (93.7500)  time: 0.9525  data: 0.7930  max mem: 17867
[05:53:06.231179] Test:  [ 500/1563]  eta: 0:02:19  loss: 1.4054 (1.3443)  acc1: 65.6250 (67.0222)  acc5: 87.5000 (90.5127)  time: 0.1301  data: 0.0002  max mem: 17867
[05:54:11.281671] Test:  [1000/1563]  eta: 0:01:13  loss: 1.6927 (1.4906)  acc1: 53.1250 (64.2295)  acc5: 87.5000 (87.8528)  time: 0.1299  data: 0.0002  max mem: 17867
[05:55:16.262938] Test:  [1500/1563]  eta: 0:00:08  loss: 1.0843 (1.5867)  acc1: 75.0000 (62.5562)  acc5: 90.6250 (86.1426)  time: 0.1299  data: 0.0002  max mem: 17867
[05:55:24.249665] Test:  [1562/1563]  eta: 0:00:00  loss: 0.6874 (1.5866)  acc1: 84.3750 (62.6020)  acc5: 93.7500 (86.1200)  time: 0.1262  data: 0.0001  max mem: 17867
[05:55:24.319219] Test: Total time: 0:03:24 (0.1305 s / it)
[05:55:24.466021] * Acc@1 62.602 Acc@5 86.120 loss 1.587
[05:55:24.466206] Accuracy of the network on the 50000 test images: 62.6%
[05:55:24.466227] Max accuracy: 64.40%
[05:55:24.490787] log_dir: ./output_dir_cml_spikformer
[05:55:25.920021] Epoch: [35]  [   0/5004]  eta: 1:59:07  lr: 0.001131  loss: 3.6731 (3.6731)  time: 1.4284  data: 0.9993  max mem: 17867
[06:07:07.447894] Epoch: [35]  [2000/5004]  eta: 0:17:35  lr: 0.001130  loss: 3.3414 (3.3353)  time: 0.3534  data: 0.0002  max mem: 17867
[06:18:47.216998] Epoch: [35]  [4000/5004]  eta: 0:05:51  lr: 0.001128  loss: 3.4684 (3.3381)  time: 0.3467  data: 0.0002  max mem: 17867
[06:24:38.332907] Epoch: [35]  [5003/5004]  eta: 0:00:00  lr: 0.001127  loss: 3.1598 (3.3402)  time: 0.3485  data: 0.0011  max mem: 17867
[06:24:38.686256] Epoch: [35] Total time: 0:29:14 (0.3506 s / it)
[06:24:38.687032] Averaged stats: lr: 0.001127  loss: 3.1598 (3.3386)
[06:24:40.014359] Test:  [   0/1563]  eta: 0:34:28  loss: 0.4828 (0.4828)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.3237  data: 1.1730  max mem: 17867
[06:25:45.172105] Test:  [ 500/1563]  eta: 0:02:21  loss: 1.3163 (1.2930)  acc1: 62.5000 (68.2011)  acc5: 93.7500 (91.4546)  time: 0.1301  data: 0.0002  max mem: 17867
[06:26:50.243639] Test:  [1000/1563]  eta: 0:01:13  loss: 2.0338 (1.4714)  acc1: 46.8750 (65.1536)  acc5: 81.2500 (88.0463)  time: 0.1300  data: 0.0002  max mem: 17867
[06:27:55.262729] Test:  [1500/1563]  eta: 0:00:08  loss: 0.8905 (1.5585)  acc1: 81.2500 (63.4056)  acc5: 93.7500 (86.4091)  time: 0.1300  data: 0.0002  max mem: 17867
[06:28:03.249370] Test:  [1562/1563]  eta: 0:00:00  loss: 0.7053 (1.5552)  acc1: 81.2500 (63.4560)  acc5: 93.7500 (86.4420)  time: 0.1263  data: 0.0001  max mem: 17867
[06:28:03.321327] Test: Total time: 0:03:24 (0.1309 s / it)
[06:28:03.447286] * Acc@1 63.456 Acc@5 86.442 loss 1.555
[06:28:03.447463] Accuracy of the network on the 50000 test images: 63.5%
[06:28:03.447486] Max accuracy: 64.40%
[06:28:03.469796] log_dir: ./output_dir_cml_spikformer
[06:28:04.910106] Epoch: [36]  [   0/5004]  eta: 2:00:02  lr: 0.001127  loss: 2.9202 (2.9202)  time: 1.4393  data: 0.8700  max mem: 17867
[06:39:44.930844] Epoch: [36]  [2000/5004]  eta: 0:17:33  lr: 0.001125  loss: 3.4070 (3.3245)  time: 0.3487  data: 0.0002  max mem: 17867
[06:51:24.839518] Epoch: [36]  [4000/5004]  eta: 0:05:51  lr: 0.001123  loss: 3.2118 (3.3296)  time: 0.3511  data: 0.0002  max mem: 17867
[06:57:15.754455] Epoch: [36]  [5003/5004]  eta: 0:00:00  lr: 0.001122  loss: 3.2265 (3.3294)  time: 0.3494  data: 0.0013  max mem: 17867
[06:57:16.125617] Epoch: [36] Total time: 0:29:12 (0.3503 s / it)
[06:57:16.126504] Averaged stats: lr: 0.001122  loss: 3.2265 (3.3287)
[06:57:17.106476] Test:  [   0/1563]  eta: 0:25:26  loss: 0.7972 (0.7972)  acc1: 75.0000 (75.0000)  acc5: 96.8750 (96.8750)  time: 0.9765  data: 0.8371  max mem: 17867
[06:58:22.253294] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.1402 (1.2263)  acc1: 65.6250 (68.5816)  acc5: 90.6250 (91.4109)  time: 0.1302  data: 0.0002  max mem: 17867
[06:59:27.285320] Test:  [1000/1563]  eta: 0:01:13  loss: 1.4052 (1.3773)  acc1: 65.6250 (66.0277)  acc5: 87.5000 (88.7706)  time: 0.1299  data: 0.0002  max mem: 17867
[07:00:32.296373] Test:  [1500/1563]  eta: 0:00:08  loss: 0.8518 (1.4765)  acc1: 81.2500 (64.2009)  acc5: 93.7500 (87.1794)  time: 0.1301  data: 0.0003  max mem: 17867
[07:00:40.284706] Test:  [1562/1563]  eta: 0:00:00  loss: 0.7228 (1.4744)  acc1: 87.5000 (64.2580)  acc5: 96.8750 (87.2140)  time: 0.1262  data: 0.0001  max mem: 17867
[07:00:40.359014] Test: Total time: 0:03:24 (0.1307 s / it)
[07:00:40.463250] * Acc@1 64.258 Acc@5 87.214 loss 1.474
[07:00:40.463456] Accuracy of the network on the 50000 test images: 64.3%
[07:00:40.463483] Max accuracy: 64.40%
[07:00:40.495350] log_dir: ./output_dir_cml_spikformer
[07:00:41.969301] Epoch: [37]  [   0/5004]  eta: 2:02:50  lr: 0.001122  loss: 3.6461 (3.6461)  time: 1.4728  data: 1.0964  max mem: 17867
[07:12:21.841962] Epoch: [37]  [2000/5004]  eta: 0:17:32  lr: 0.001120  loss: 3.2218 (3.3196)  time: 0.3498  data: 0.0002  max mem: 17867
[07:24:01.660348] Epoch: [37]  [4000/5004]  eta: 0:05:51  lr: 0.001118  loss: 3.2667 (3.3183)  time: 0.3464  data: 0.0002  max mem: 17867
[07:29:52.514781] Epoch: [37]  [5003/5004]  eta: 0:00:00  lr: 0.001117  loss: 3.3820 (3.3222)  time: 0.3465  data: 0.0006  max mem: 17867
[07:29:52.854327] Epoch: [37] Total time: 0:29:12 (0.3502 s / it)
[07:29:52.864444] Averaged stats: lr: 0.001117  loss: 3.3820 (3.3240)
[07:29:53.813854] Test:  [   0/1563]  eta: 0:24:36  loss: 0.7710 (0.7710)  acc1: 81.2500 (81.2500)  acc5: 93.7500 (93.7500)  time: 0.9447  data: 0.7989  max mem: 17867
[07:30:58.972686] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.3044 (1.2375)  acc1: 65.6250 (67.8331)  acc5: 93.7500 (91.5669)  time: 0.1300  data: 0.0002  max mem: 17867
[07:32:03.965354] Test:  [1000/1563]  eta: 0:01:13  loss: 1.6506 (1.3948)  acc1: 59.3750 (65.3378)  acc5: 87.5000 (88.6957)  time: 0.1300  data: 0.0002  max mem: 17867
[07:33:08.962291] Test:  [1500/1563]  eta: 0:00:08  loss: 1.0170 (1.4872)  acc1: 75.0000 (63.5097)  acc5: 90.6250 (87.1669)  time: 0.1301  data: 0.0002  max mem: 17867
[07:33:16.948070] Test:  [1562/1563]  eta: 0:00:00  loss: 0.7163 (1.4870)  acc1: 81.2500 (63.5740)  acc5: 93.7500 (87.1460)  time: 0.1263  data: 0.0001  max mem: 17867
[07:33:17.161524] Test: Total time: 0:03:24 (0.1307 s / it)
[07:33:17.172888] * Acc@1 63.574 Acc@5 87.146 loss 1.487
[07:33:17.173019] Accuracy of the network on the 50000 test images: 63.6%
[07:33:17.173041] Max accuracy: 64.40%
[07:33:17.194848] log_dir: ./output_dir_cml_spikformer
[07:33:18.803843] Epoch: [38]  [   0/5004]  eta: 2:14:08  lr: 0.001117  loss: 3.6652 (3.6652)  time: 1.6083  data: 1.1187  max mem: 17867
[07:44:59.120114] Epoch: [38]  [2000/5004]  eta: 0:17:33  lr: 0.001115  loss: 3.1310 (3.3057)  time: 0.3513  data: 0.0002  max mem: 17867
[07:56:39.377536] Epoch: [38]  [4000/5004]  eta: 0:05:51  lr: 0.001113  loss: 3.2943 (3.3096)  time: 0.3490  data: 0.0002  max mem: 17867
[08:02:30.426423] Epoch: [38]  [5003/5004]  eta: 0:00:00  lr: 0.001112  loss: 3.2954 (3.3116)  time: 0.3462  data: 0.0011  max mem: 17867
[08:02:30.777862] Epoch: [38] Total time: 0:29:13 (0.3504 s / it)
[08:02:30.788492] Averaged stats: lr: 0.001112  loss: 3.2954 (3.3147)
[08:02:31.773318] Test:  [   0/1563]  eta: 0:25:33  loss: 0.6534 (0.6534)  acc1: 87.5000 (87.5000)  acc5: 90.6250 (90.6250)  time: 0.9813  data: 0.8156  max mem: 17867
[08:03:36.961623] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.2673 (1.2160)  acc1: 68.7500 (69.0494)  acc5: 93.7500 (91.9536)  time: 0.1305  data: 0.0002  max mem: 17867
[08:04:42.011362] Test:  [1000/1563]  eta: 0:01:13  loss: 1.6072 (1.3900)  acc1: 56.2500 (66.2369)  acc5: 87.5000 (89.0828)  time: 0.1299  data: 0.0002  max mem: 17867
[08:05:46.999110] Test:  [1500/1563]  eta: 0:00:08  loss: 1.1210 (1.5069)  acc1: 75.0000 (63.8928)  acc5: 90.6250 (87.1273)  time: 0.1299  data: 0.0002  max mem: 17867
[08:05:54.982510] Test:  [1562/1563]  eta: 0:00:00  loss: 0.5711 (1.4987)  acc1: 87.5000 (64.0080)  acc5: 96.8750 (87.2560)  time: 0.1262  data: 0.0001  max mem: 17867
[08:05:55.057380] Test: Total time: 0:03:24 (0.1307 s / it)
[08:05:55.298791] * Acc@1 64.008 Acc@5 87.256 loss 1.499
[08:05:55.298942] Accuracy of the network on the 50000 test images: 64.0%
[08:05:55.298961] Max accuracy: 64.40%
[08:05:55.338916] log_dir: ./output_dir_cml_spikformer
[08:05:56.900146] Epoch: [39]  [   0/5004]  eta: 2:10:07  lr: 0.001112  loss: 3.7324 (3.7324)  time: 1.5603  data: 1.0502  max mem: 17867
[08:17:37.572075] Epoch: [39]  [2000/5004]  eta: 0:17:34  lr: 0.001110  loss: 3.3073 (3.3169)  time: 0.3481  data: 0.0002  max mem: 17867
[08:29:17.867342] Epoch: [39]  [4000/5004]  eta: 0:05:51  lr: 0.001108  loss: 3.2403 (3.3158)  time: 0.3469  data: 0.0002  max mem: 17867
[08:35:09.183441] Epoch: [39]  [5003/5004]  eta: 0:00:00  lr: 0.001107  loss: 3.4998 (3.3155)  time: 0.3464  data: 0.0011  max mem: 17867
[08:35:09.527498] Epoch: [39] Total time: 0:29:14 (0.3506 s / it)
[08:35:09.532825] Averaged stats: lr: 0.001107  loss: 3.4998 (3.3102)
[08:35:10.743907] Test:  [   0/1563]  eta: 0:31:27  loss: 1.0253 (1.0253)  acc1: 81.2500 (81.2500)  acc5: 93.7500 (93.7500)  time: 1.2076  data: 1.0707  max mem: 17867
[08:36:15.719493] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.4221 (1.3125)  acc1: 65.6250 (67.4900)  acc5: 90.6250 (90.9993)  time: 0.1299  data: 0.0002  max mem: 17867
[08:37:20.668219] Test:  [1000/1563]  eta: 0:01:13  loss: 1.6839 (1.4630)  acc1: 56.2500 (64.9632)  acc5: 90.6250 (88.2524)  time: 0.1299  data: 0.0002  max mem: 17867
[08:38:25.713607] Test:  [1500/1563]  eta: 0:00:08  loss: 0.9502 (1.5593)  acc1: 75.0000 (63.0600)  acc5: 93.7500 (86.6797)  time: 0.1301  data: 0.0002  max mem: 17867
[08:38:33.695237] Test:  [1562/1563]  eta: 0:00:00  loss: 0.8619 (1.5623)  acc1: 87.5000 (63.0540)  acc5: 93.7500 (86.6280)  time: 0.1262  data: 0.0001  max mem: 17867
[08:38:33.764249] Test: Total time: 0:03:24 (0.1307 s / it)
[08:38:33.931230] * Acc@1 63.054 Acc@5 86.628 loss 1.562
[08:38:33.931409] Accuracy of the network on the 50000 test images: 63.1%
[08:38:33.931430] Max accuracy: 64.40%
[08:38:33.953349] log_dir: ./output_dir_cml_spikformer
[08:38:35.565054] Epoch: [40]  [   0/5004]  eta: 2:14:21  lr: 0.001107  loss: 3.4931 (3.4931)  time: 1.6109  data: 0.9734  max mem: 17867
[08:50:15.585266] Epoch: [40]  [2000/5004]  eta: 0:17:33  lr: 0.001105  loss: 3.3107 (3.2990)  time: 0.3522  data: 0.0002  max mem: 17867
[09:01:55.086864] Epoch: [40]  [4000/5004]  eta: 0:05:51  lr: 0.001103  loss: 3.2018 (3.3017)  time: 0.3477  data: 0.0002  max mem: 17867
[09:07:46.206615] Epoch: [40]  [5003/5004]  eta: 0:00:00  lr: 0.001102  loss: 3.1398 (3.3082)  time: 0.3493  data: 0.0011  max mem: 17867
[09:07:46.572684] Epoch: [40] Total time: 0:29:12 (0.3502 s / it)
[09:07:46.573489] Averaged stats: lr: 0.001102  loss: 3.1398 (3.3024)
[09:07:47.630166] Test:  [   0/1563]  eta: 0:27:24  loss: 0.4143 (0.4143)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.0519  data: 0.9051  max mem: 17867
[09:08:52.739484] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.0530 (1.1547)  acc1: 68.7500 (70.1722)  acc5: 90.6250 (92.2967)  time: 0.1311  data: 0.0002  max mem: 17867
[09:09:57.940011] Test:  [1000/1563]  eta: 0:01:13  loss: 2.1717 (1.3212)  acc1: 46.8750 (67.1735)  acc5: 78.1250 (89.5042)  time: 0.1299  data: 0.0002  max mem: 17867
[09:11:02.917580] Test:  [1500/1563]  eta: 0:00:08  loss: 0.8819 (1.4303)  acc1: 81.2500 (65.1961)  acc5: 93.7500 (87.7353)  time: 0.1299  data: 0.0002  max mem: 17867
[09:11:10.900486] Test:  [1562/1563]  eta: 0:00:00  loss: 0.6732 (1.4297)  acc1: 81.2500 (65.1920)  acc5: 93.7500 (87.7520)  time: 0.1262  data: 0.0001  max mem: 17867
[09:11:10.967818] Test: Total time: 0:03:24 (0.1308 s / it)
[09:11:10.968953] * Acc@1 65.192 Acc@5 87.752 loss 1.430
[09:11:10.969072] Accuracy of the network on the 50000 test images: 65.2%
[09:11:10.969094] Max accuracy: 65.19%
[09:11:10.993346] log_dir: ./output_dir_cml_spikformer
[09:11:12.403633] Epoch: [41]  [   0/5004]  eta: 1:57:33  lr: 0.001102  loss: 2.8559 (2.8559)  time: 1.4096  data: 1.0323  max mem: 17867
[09:22:52.421724] Epoch: [41]  [2000/5004]  eta: 0:17:32  lr: 0.001100  loss: 3.2336 (3.2803)  time: 0.3522  data: 0.0002  max mem: 17867
[09:34:32.308948] Epoch: [41]  [4000/5004]  eta: 0:05:51  lr: 0.001098  loss: 3.2662 (3.2940)  time: 0.3486  data: 0.0002  max mem: 17867
[09:40:23.700645] Epoch: [41]  [5003/5004]  eta: 0:00:00  lr: 0.001097  loss: 3.2208 (3.2967)  time: 0.3483  data: 0.0011  max mem: 17867
[09:40:24.051988] Epoch: [41] Total time: 0:29:13 (0.3503 s / it)
[09:40:24.053896] Averaged stats: lr: 0.001097  loss: 3.2208 (3.2997)
[09:40:25.074512] Test:  [   0/1563]  eta: 0:26:29  loss: 0.7199 (0.7199)  acc1: 84.3750 (84.3750)  acc5: 96.8750 (96.8750)  time: 1.0172  data: 0.8761  max mem: 17867
[09:41:30.117484] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.0686 (1.2004)  acc1: 75.0000 (69.7480)  acc5: 93.7500 (92.0409)  time: 0.1298  data: 0.0002  max mem: 17867
[09:42:35.116103] Test:  [1000/1563]  eta: 0:01:13  loss: 1.4756 (1.3490)  acc1: 68.7500 (67.1391)  acc5: 90.6250 (89.3981)  time: 0.1299  data: 0.0002  max mem: 17867
[09:43:40.116542] Test:  [1500/1563]  eta: 0:00:08  loss: 0.8522 (1.4527)  acc1: 78.1250 (64.9421)  acc5: 93.7500 (87.7998)  time: 0.1303  data: 0.0002  max mem: 17867
[09:43:48.099670] Test:  [1562/1563]  eta: 0:00:00  loss: 0.7497 (1.4544)  acc1: 87.5000 (64.9140)  acc5: 93.7500 (87.8040)  time: 0.1262  data: 0.0001  max mem: 17867
[09:43:48.172488] Test: Total time: 0:03:24 (0.1306 s / it)
[09:43:48.596194] * Acc@1 64.914 Acc@5 87.804 loss 1.454
[09:43:48.596342] Accuracy of the network on the 50000 test images: 64.9%
[09:43:48.596362] Max accuracy: 65.19%
[09:43:48.604219] log_dir: ./output_dir_cml_spikformer
[09:43:50.170729] Epoch: [42]  [   0/5004]  eta: 2:10:35  lr: 0.001097  loss: 3.2898 (3.2898)  time: 1.5658  data: 0.9297  max mem: 17867
[09:55:30.269448] Epoch: [42]  [2000/5004]  eta: 0:17:33  lr: 0.001094  loss: 3.3183 (3.2827)  time: 0.3529  data: 0.0002  max mem: 17867
[10:07:09.515019] Epoch: [42]  [4000/5004]  eta: 0:05:51  lr: 0.001092  loss: 3.1600 (3.2902)  time: 0.3468  data: 0.0002  max mem: 17867
[10:13:00.216664] Epoch: [42]  [5003/5004]  eta: 0:00:00  lr: 0.001091  loss: 3.0862 (3.2899)  time: 0.3437  data: 0.0011  max mem: 17867
[10:13:00.593687] Epoch: [42] Total time: 0:29:11 (0.3501 s / it)
[10:13:00.601365] Averaged stats: lr: 0.001091  loss: 3.0862 (3.2912)
[10:13:01.592668] Test:  [   0/1563]  eta: 0:25:43  loss: 0.4963 (0.4963)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 0.9876  data: 0.8434  max mem: 17867
[10:14:06.661618] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.1132 (1.2033)  acc1: 71.8750 (69.4174)  acc5: 90.6250 (91.6105)  time: 0.1300  data: 0.0002  max mem: 17867
[10:15:11.608505] Test:  [1000/1563]  eta: 0:01:13  loss: 1.6465 (1.3761)  acc1: 62.5000 (66.1776)  acc5: 90.6250 (88.5864)  time: 0.1299  data: 0.0002  max mem: 17867
[10:16:16.609204] Test:  [1500/1563]  eta: 0:00:08  loss: 0.8002 (1.4641)  acc1: 81.2500 (64.4903)  acc5: 93.7500 (87.0899)  time: 0.1298  data: 0.0002  max mem: 17867
[10:16:24.587518] Test:  [1562/1563]  eta: 0:00:00  loss: 0.5537 (1.4583)  acc1: 87.5000 (64.6420)  acc5: 96.8750 (87.1600)  time: 0.1261  data: 0.0001  max mem: 17867
[10:16:24.651892] Test: Total time: 0:03:24 (0.1305 s / it)
[10:16:24.974816] * Acc@1 64.642 Acc@5 87.160 loss 1.458
[10:16:24.974957] Accuracy of the network on the 50000 test images: 64.6%
[10:16:24.974977] Max accuracy: 65.19%
[10:16:25.018356] log_dir: ./output_dir_cml_spikformer
[10:16:26.538433] Epoch: [43]  [   0/5004]  eta: 2:06:40  lr: 0.001091  loss: 3.2339 (3.2339)  time: 1.5189  data: 0.9828  max mem: 17867
[10:28:05.444085] Epoch: [43]  [2000/5004]  eta: 0:17:31  lr: 0.001089  loss: 3.3811 (3.2955)  time: 0.3545  data: 0.0002  max mem: 17867
[10:39:44.840329] Epoch: [43]  [4000/5004]  eta: 0:05:51  lr: 0.001087  loss: 3.3076 (3.2856)  time: 0.3468  data: 0.0002  max mem: 17867
[10:45:35.494853] Epoch: [43]  [5003/5004]  eta: 0:00:00  lr: 0.001086  loss: 3.2191 (3.2846)  time: 0.3451  data: 0.0011  max mem: 17867
[10:45:35.877594] Epoch: [43] Total time: 0:29:10 (0.3499 s / it)
[10:45:35.890186] Averaged stats: lr: 0.001086  loss: 3.2191 (3.2839)
[10:45:36.872484] Test:  [   0/1563]  eta: 0:25:29  loss: 0.6486 (0.6486)  acc1: 87.5000 (87.5000)  acc5: 96.8750 (96.8750)  time: 0.9786  data: 0.8384  max mem: 17867
[10:46:41.851466] Test:  [ 500/1563]  eta: 0:02:19  loss: 0.9414 (1.2071)  acc1: 71.8750 (69.5796)  acc5: 93.7500 (91.9224)  time: 0.1299  data: 0.0002  max mem: 17867
[10:47:46.846449] Test:  [1000/1563]  eta: 0:01:13  loss: 1.5398 (1.3465)  acc1: 62.5000 (67.0111)  acc5: 87.5000 (89.5042)  time: 0.1299  data: 0.0002  max mem: 17867
[10:48:51.848933] Test:  [1500/1563]  eta: 0:00:08  loss: 0.8417 (1.4463)  acc1: 81.2500 (64.9484)  acc5: 93.7500 (87.9476)  time: 0.1300  data: 0.0002  max mem: 17867
[10:48:59.924342] Test:  [1562/1563]  eta: 0:00:00  loss: 0.7401 (1.4457)  acc1: 84.3750 (64.9860)  acc5: 96.8750 (87.9700)  time: 0.1307  data: 0.0001  max mem: 17867
[10:49:00.021674] Test: Total time: 0:03:24 (0.1306 s / it)
[10:49:00.263932] * Acc@1 64.986 Acc@5 87.970 loss 1.446
[10:49:00.264104] Accuracy of the network on the 50000 test images: 65.0%
[10:49:00.264128] Max accuracy: 65.19%
[10:49:00.277581] log_dir: ./output_dir_cml_spikformer
[10:49:01.767067] Epoch: [44]  [   0/5004]  eta: 2:04:09  lr: 0.001086  loss: 2.8007 (2.8007)  time: 1.4887  data: 1.0058  max mem: 17867
[11:00:42.444773] Epoch: [44]  [2000/5004]  eta: 0:17:34  lr: 0.001083  loss: 3.4771 (3.2619)  time: 0.3476  data: 0.0002  max mem: 17867
[11:12:21.591924] Epoch: [44]  [4000/5004]  eta: 0:05:51  lr: 0.001081  loss: 3.3839 (3.2680)  time: 0.3501  data: 0.0002  max mem: 17867
[11:18:12.018068] Epoch: [44]  [5003/5004]  eta: 0:00:00  lr: 0.001080  loss: 3.2811 (3.2685)  time: 0.3437  data: 0.0011  max mem: 17867
[11:18:12.498367] Epoch: [44] Total time: 0:29:12 (0.3502 s / it)
[11:18:12.524943] Averaged stats: lr: 0.001080  loss: 3.2811 (3.2762)
[11:18:14.383711] Test:  [   0/1563]  eta: 0:48:19  loss: 0.7277 (0.7277)  acc1: 90.6250 (90.6250)  acc5: 96.8750 (96.8750)  time: 1.8553  data: 1.6029  max mem: 17867
[11:19:19.351700] Test:  [ 500/1563]  eta: 0:02:21  loss: 1.1986 (1.2232)  acc1: 65.6250 (69.4923)  acc5: 90.6250 (92.0409)  time: 0.1298  data: 0.0002  max mem: 17867
[11:20:24.322263] Test:  [1000/1563]  eta: 0:01:14  loss: 1.5592 (1.3816)  acc1: 59.3750 (67.0704)  acc5: 90.6250 (89.4231)  time: 0.1299  data: 0.0002  max mem: 17867
[11:21:29.305833] Test:  [1500/1563]  eta: 0:00:08  loss: 1.0020 (1.4831)  acc1: 78.1250 (64.9484)  acc5: 93.7500 (87.8373)  time: 0.1299  data: 0.0002  max mem: 17867
[11:21:37.312156] Test:  [1562/1563]  eta: 0:00:00  loss: 0.9174 (1.4845)  acc1: 84.3750 (64.9960)  acc5: 96.8750 (87.7960)  time: 0.1270  data: 0.0001  max mem: 17867
[11:21:37.378679] Test: Total time: 0:03:24 (0.1311 s / it)
[11:21:37.570883] * Acc@1 64.996 Acc@5 87.796 loss 1.485
[11:21:37.571058] Accuracy of the network on the 50000 test images: 65.0%
[11:21:37.571083] Max accuracy: 65.19%
[11:21:37.617042] log_dir: ./output_dir_cml_spikformer
[11:21:39.147457] Epoch: [45]  [   0/5004]  eta: 2:07:35  lr: 0.001080  loss: 3.1099 (3.1099)  time: 1.5298  data: 0.9989  max mem: 17867
[11:33:19.559049] Epoch: [45]  [2000/5004]  eta: 0:17:33  lr: 0.001077  loss: 3.2391 (3.2732)  time: 0.3560  data: 0.0003  max mem: 17867
[11:44:58.632979] Epoch: [45]  [4000/5004]  eta: 0:05:51  lr: 0.001075  loss: 3.2441 (3.2777)  time: 0.3493  data: 0.0002  max mem: 17867
[11:50:49.160563] Epoch: [45]  [5003/5004]  eta: 0:00:00  lr: 0.001074  loss: 3.2100 (3.2769)  time: 0.3450  data: 0.0011  max mem: 17867
[11:50:49.509310] Epoch: [45] Total time: 0:29:11 (0.3501 s / it)
[11:50:49.510076] Averaged stats: lr: 0.001074  loss: 3.2100 (3.2729)
[11:50:50.484008] Test:  [   0/1563]  eta: 0:25:16  loss: 0.6989 (0.6989)  acc1: 87.5000 (87.5000)  acc5: 96.8750 (96.8750)  time: 0.9705  data: 0.7964  max mem: 17867
[11:51:55.558513] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.3075 (1.1914)  acc1: 65.6250 (69.7667)  acc5: 87.5000 (91.8288)  time: 0.1307  data: 0.0002  max mem: 17867
[11:53:00.786960] Test:  [1000/1563]  eta: 0:01:13  loss: 1.5633 (1.3594)  acc1: 56.2500 (66.4585)  acc5: 90.6250 (89.3076)  time: 0.1312  data: 0.0002  max mem: 17867
[11:54:05.781466] Test:  [1500/1563]  eta: 0:00:08  loss: 0.7914 (1.4571)  acc1: 78.1250 (64.6277)  acc5: 93.7500 (87.6395)  time: 0.1299  data: 0.0002  max mem: 17867
[11:54:13.763902] Test:  [1562/1563]  eta: 0:00:00  loss: 0.7797 (1.4551)  acc1: 84.3750 (64.6960)  acc5: 93.7500 (87.6760)  time: 0.1262  data: 0.0001  max mem: 17867
[11:54:13.821727] Test: Total time: 0:03:24 (0.1307 s / it)
[11:54:13.942452] * Acc@1 64.696 Acc@5 87.676 loss 1.455
[11:54:13.942600] Accuracy of the network on the 50000 test images: 64.7%
[11:54:13.942622] Max accuracy: 65.19%
[11:54:13.965027] log_dir: ./output_dir_cml_spikformer
[11:54:15.548840] Epoch: [46]  [   0/5004]  eta: 2:12:01  lr: 0.001074  loss: 2.9909 (2.9909)  time: 1.5831  data: 0.8941  max mem: 17867
[12:05:56.190788] Epoch: [46]  [2000/5004]  eta: 0:17:34  lr: 0.001072  loss: 3.3052 (3.2628)  time: 0.3500  data: 0.0002  max mem: 17867
[12:17:35.226148] Epoch: [46]  [4000/5004]  eta: 0:05:51  lr: 0.001069  loss: 3.3193 (3.2664)  time: 0.3510  data: 0.0002  max mem: 17867
[12:23:25.393760] Epoch: [46]  [5003/5004]  eta: 0:00:00  lr: 0.001068  loss: 3.1880 (3.2704)  time: 0.3440  data: 0.0006  max mem: 17867
[12:23:25.779342] Epoch: [46] Total time: 0:29:11 (0.3501 s / it)
[12:23:25.787967] Averaged stats: lr: 0.001068  loss: 3.1880 (3.2633)
[12:23:26.758770] Test:  [   0/1563]  eta: 0:25:11  loss: 0.5087 (0.5087)  acc1: 90.6250 (90.6250)  acc5: 96.8750 (96.8750)  time: 0.9673  data: 0.8223  max mem: 17867
[12:24:31.754548] Test:  [ 500/1563]  eta: 0:02:19  loss: 1.0646 (1.1592)  acc1: 68.7500 (70.0536)  acc5: 93.7500 (92.0409)  time: 0.1299  data: 0.0002  max mem: 17867
[12:25:36.752408] Test:  [1000/1563]  eta: 0:01:13  loss: 1.5932 (1.3395)  acc1: 59.3750 (66.8925)  acc5: 81.2500 (89.2202)  time: 0.1298  data: 0.0002  max mem: 17867
[12:26:41.709979] Test:  [1500/1563]  eta: 0:00:08  loss: 0.7593 (1.4407)  acc1: 78.1250 (65.1566)  acc5: 93.7500 (87.7519)  time: 0.1300  data: 0.0002  max mem: 17867
[12:26:49.695122] Test:  [1562/1563]  eta: 0:00:00  loss: 0.6946 (1.4391)  acc1: 84.3750 (65.1980)  acc5: 96.8750 (87.7740)  time: 0.1261  data: 0.0001  max mem: 17867
[12:26:49.753212] Test: Total time: 0:03:23 (0.1305 s / it)
[12:26:50.046716] * Acc@1 65.198 Acc@5 87.774 loss 1.439
[12:26:50.046961] Accuracy of the network on the 50000 test images: 65.2%
[12:26:50.046984] Max accuracy: 65.20%
[12:26:50.066962] log_dir: ./output_dir_cml_spikformer
[12:26:51.554355] Epoch: [47]  [   0/5004]  eta: 2:03:59  lr: 0.001068  loss: 2.7178 (2.7178)  time: 1.4867  data: 1.1213  max mem: 17867
[12:38:31.037776] Epoch: [47]  [2000/5004]  eta: 0:17:32  lr: 0.001065  loss: 3.3464 (3.2495)  time: 0.3501  data: 0.0002  max mem: 17867
[12:50:10.140322] Epoch: [47]  [4000/5004]  eta: 0:05:51  lr: 0.001063  loss: 3.3806 (3.2594)  time: 0.3468  data: 0.0003  max mem: 17867
[12:56:00.339749] Epoch: [47]  [5003/5004]  eta: 0:00:00  lr: 0.001062  loss: 3.2183 (3.2575)  time: 0.3439  data: 0.0006  max mem: 17867
[12:56:00.740922] Epoch: [47] Total time: 0:29:10 (0.3499 s / it)
[12:56:00.741604] Averaged stats: lr: 0.001062  loss: 3.2183 (3.2609)
[12:56:01.780348] Test:  [   0/1563]  eta: 0:26:58  loss: 0.4136 (0.4136)  acc1: 90.6250 (90.6250)  acc5: 96.8750 (96.8750)  time: 1.0353  data: 0.8863  max mem: 17867
[12:57:06.811373] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.2908 (1.1801)  acc1: 59.3750 (70.2345)  acc5: 93.7500 (92.1220)  time: 0.1299  data: 0.0002  max mem: 17867
[12:58:11.801801] Test:  [1000/1563]  eta: 0:01:13  loss: 1.5842 (1.3313)  acc1: 53.1250 (67.2890)  acc5: 87.5000 (89.5011)  time: 0.1301  data: 0.0004  max mem: 17867
[12:59:16.798398] Test:  [1500/1563]  eta: 0:00:08  loss: 0.8526 (1.4357)  acc1: 81.2500 (65.1087)  acc5: 90.6250 (87.8914)  time: 0.1299  data: 0.0002  max mem: 17867
[12:59:24.783064] Test:  [1562/1563]  eta: 0:00:00  loss: 0.7261 (1.4327)  acc1: 84.3750 (65.1800)  acc5: 93.7500 (87.9220)  time: 0.1262  data: 0.0001  max mem: 17867
[12:59:24.851001] Test: Total time: 0:03:24 (0.1306 s / it)
[12:59:25.093394] * Acc@1 65.180 Acc@5 87.922 loss 1.433
[12:59:25.093540] Accuracy of the network on the 50000 test images: 65.2%
[12:59:25.093560] Max accuracy: 65.20%
[12:59:25.133450] log_dir: ./output_dir_cml_spikformer
[12:59:26.637684] Epoch: [48]  [   0/5004]  eta: 2:05:20  lr: 0.001062  loss: 3.2558 (3.2558)  time: 1.5029  data: 1.1005  max mem: 17867
[13:11:06.951559] Epoch: [48]  [2000/5004]  eta: 0:17:33  lr: 0.001059  loss: 3.2572 (3.2470)  time: 0.3497  data: 0.0002  max mem: 17867
[13:22:46.178153] Epoch: [48]  [4000/5004]  eta: 0:05:51  lr: 0.001057  loss: 3.0384 (3.2538)  time: 0.3464  data: 0.0002  max mem: 17867
[13:28:36.820965] Epoch: [48]  [5003/5004]  eta: 0:00:00  lr: 0.001056  loss: 3.3014 (3.2573)  time: 0.3479  data: 0.0012  max mem: 17867
[13:28:37.183603] Epoch: [48] Total time: 0:29:12 (0.3501 s / it)
[13:28:37.184379] Averaged stats: lr: 0.001056  loss: 3.3014 (3.2588)
[13:28:38.224917] Test:  [   0/1563]  eta: 0:27:00  loss: 0.2565 (0.2565)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.0368  data: 0.8953  max mem: 17867
[13:29:43.248231] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.2461 (1.1836)  acc1: 71.8750 (70.3530)  acc5: 87.5000 (92.4588)  time: 0.1299  data: 0.0002  max mem: 17867
[13:30:48.256532] Test:  [1000/1563]  eta: 0:01:13  loss: 1.5380 (1.3648)  acc1: 62.5000 (66.9018)  acc5: 90.6250 (89.3919)  time: 0.1299  data: 0.0002  max mem: 17867
[13:31:53.297298] Test:  [1500/1563]  eta: 0:00:08  loss: 0.8312 (1.4645)  acc1: 81.2500 (64.9567)  acc5: 93.7500 (87.7249)  time: 0.1299  data: 0.0002  max mem: 17867
[13:32:01.281828] Test:  [1562/1563]  eta: 0:00:00  loss: 0.7621 (1.4596)  acc1: 87.5000 (65.0860)  acc5: 93.7500 (87.8100)  time: 0.1263  data: 0.0001  max mem: 17867
[13:32:01.357686] Test: Total time: 0:03:24 (0.1306 s / it)
[13:32:01.608209] * Acc@1 65.086 Acc@5 87.810 loss 1.460
[13:32:01.608460] Accuracy of the network on the 50000 test images: 65.1%
[13:32:01.608482] Max accuracy: 65.20%
[13:32:01.635732] log_dir: ./output_dir_cml_spikformer
[13:32:03.297188] Epoch: [49]  [   0/5004]  eta: 2:18:29  lr: 0.001056  loss: 3.1192 (3.1192)  time: 1.6606  data: 1.1992  max mem: 17867
[13:43:42.241475] Epoch: [49]  [2000/5004]  eta: 0:17:31  lr: 0.001053  loss: 3.1471 (3.2439)  time: 0.3481  data: 0.0002  max mem: 17867
[13:55:20.571273] Epoch: [49]  [4000/5004]  eta: 0:05:51  lr: 0.001051  loss: 3.2028 (3.2547)  time: 0.3481  data: 0.0002  max mem: 17867
[14:01:10.449528] Epoch: [49]  [5003/5004]  eta: 0:00:00  lr: 0.001049  loss: 3.2322 (3.2551)  time: 0.3483  data: 0.0011  max mem: 17867
[14:01:10.806117] Epoch: [49] Total time: 0:29:09 (0.3496 s / it)
[14:01:10.806995] Averaged stats: lr: 0.001049  loss: 3.2322 (3.2500)
[14:01:11.814888] Test:  [   0/1563]  eta: 0:26:04  loss: 0.3797 (0.3797)  acc1: 90.6250 (90.6250)  acc5: 96.8750 (96.8750)  time: 1.0012  data: 0.8628  max mem: 17867
[14:02:16.836469] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.1113 (1.1082)  acc1: 71.8750 (70.8146)  acc5: 93.7500 (92.8081)  time: 0.1301  data: 0.0002  max mem: 17867
[14:03:21.910793] Test:  [1000/1563]  eta: 0:01:13  loss: 1.6402 (1.2773)  acc1: 56.2500 (67.6261)  acc5: 87.5000 (90.2223)  time: 0.1302  data: 0.0002  max mem: 17867
[14:04:26.892794] Test:  [1500/1563]  eta: 0:00:08  loss: 0.6402 (1.3925)  acc1: 81.2500 (65.5625)  acc5: 96.8750 (88.3494)  time: 0.1299  data: 0.0002  max mem: 17867
[14:04:34.873095] Test:  [1562/1563]  eta: 0:00:00  loss: 0.7027 (1.3936)  acc1: 84.3750 (65.5900)  acc5: 93.7500 (88.3460)  time: 0.1262  data: 0.0001  max mem: 17867
[14:04:34.934556] Test: Total time: 0:03:24 (0.1306 s / it)
[14:04:35.168962] * Acc@1 65.590 Acc@5 88.346 loss 1.394
[14:04:35.169126] Accuracy of the network on the 50000 test images: 65.6%
[14:04:35.169153] Max accuracy: 65.59%
[14:04:35.192963] log_dir: ./output_dir_cml_spikformer
[14:04:36.667476] Epoch: [50]  [   0/5004]  eta: 2:02:53  lr: 0.001049  loss: 3.8998 (3.8998)  time: 1.4734  data: 1.0390  max mem: 17867
[14:16:15.732329] Epoch: [50]  [2000/5004]  eta: 0:17:31  lr: 0.001047  loss: 3.1246 (3.2354)  time: 0.3530  data: 0.0002  max mem: 17867
[14:27:54.311591] Epoch: [50]  [4000/5004]  eta: 0:05:51  lr: 0.001044  loss: 3.2599 (3.2446)  time: 0.3446  data: 0.0002  max mem: 17867
[14:33:44.899955] Epoch: [50]  [5003/5004]  eta: 0:00:00  lr: 0.001043  loss: 3.1574 (3.2491)  time: 0.3444  data: 0.0012  max mem: 17867
[14:33:45.253659] Epoch: [50] Total time: 0:29:10 (0.3497 s / it)
[14:33:45.254936] Averaged stats: lr: 0.001043  loss: 3.1574 (3.2460)
[14:33:46.315483] Test:  [   0/1563]  eta: 0:27:32  loss: 0.5488 (0.5488)  acc1: 90.6250 (90.6250)  acc5: 96.8750 (96.8750)  time: 1.0570  data: 0.9221  max mem: 17867
[14:34:51.319986] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.2645 (1.1683)  acc1: 71.8750 (70.7897)  acc5: 93.7500 (92.5274)  time: 0.1300  data: 0.0002  max mem: 17867
[14:35:56.341974] Test:  [1000/1563]  eta: 0:01:13  loss: 1.4830 (1.3348)  acc1: 56.2500 (67.6324)  acc5: 90.6250 (89.6916)  time: 0.1299  data: 0.0002  max mem: 17867
[14:37:01.332110] Test:  [1500/1563]  eta: 0:00:08  loss: 0.9036 (1.4222)  acc1: 81.2500 (65.9623)  acc5: 93.7500 (88.2995)  time: 0.1299  data: 0.0002  max mem: 17867
[14:37:09.315503] Test:  [1562/1563]  eta: 0:00:00  loss: 0.7141 (1.4215)  acc1: 87.5000 (66.0060)  acc5: 93.7500 (88.3340)  time: 0.1262  data: 0.0001  max mem: 17867
[14:37:09.379889] Test: Total time: 0:03:24 (0.1306 s / it)
[14:37:09.594314] * Acc@1 66.006 Acc@5 88.334 loss 1.421
[14:37:09.594453] Accuracy of the network on the 50000 test images: 66.0%
[14:37:09.594474] Max accuracy: 66.01%
[14:37:09.620122] log_dir: ./output_dir_cml_spikformer
[14:37:11.052111] Epoch: [51]  [   0/5004]  eta: 1:59:22  lr: 0.001043  loss: 3.4588 (3.4588)  time: 1.4313  data: 0.9305  max mem: 17867
[14:48:50.607212] Epoch: [51]  [2000/5004]  eta: 0:17:32  lr: 0.001040  loss: 3.1961 (3.2544)  time: 0.3496  data: 0.0002  max mem: 17867
[15:00:30.232065] Epoch: [51]  [4000/5004]  eta: 0:05:51  lr: 0.001038  loss: 3.2658 (3.2457)  time: 0.3529  data: 0.0002  max mem: 17867
[15:06:20.913818] Epoch: [51]  [5003/5004]  eta: 0:00:00  lr: 0.001036  loss: 3.0910 (3.2421)  time: 0.3437  data: 0.0011  max mem: 17867
[15:06:21.282015] Epoch: [51] Total time: 0:29:11 (0.3501 s / it)
[15:06:21.289152] Averaged stats: lr: 0.001036  loss: 3.0910 (3.2412)
[15:06:22.387559] Test:  [   0/1563]  eta: 0:28:31  loss: 0.3041 (0.3041)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.0948  data: 0.9276  max mem: 17867
[15:07:27.423461] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.8367 (1.1495)  acc1: 75.0000 (70.1534)  acc5: 96.8750 (92.4775)  time: 0.1300  data: 0.0002  max mem: 17867
[15:08:32.373634] Test:  [1000/1563]  eta: 0:01:13  loss: 1.2235 (1.3059)  acc1: 62.5000 (67.3701)  acc5: 93.7500 (89.8664)  time: 0.1298  data: 0.0002  max mem: 17867
[15:09:37.366467] Test:  [1500/1563]  eta: 0:00:08  loss: 0.8109 (1.4008)  acc1: 81.2500 (65.6625)  acc5: 93.7500 (88.2891)  time: 0.1298  data: 0.0002  max mem: 17867
[15:09:45.350232] Test:  [1562/1563]  eta: 0:00:00  loss: 0.6411 (1.3996)  acc1: 81.2500 (65.6900)  acc5: 93.7500 (88.2780)  time: 0.1264  data: 0.0001  max mem: 17867
[15:09:45.419986] Test: Total time: 0:03:24 (0.1306 s / it)
[15:09:45.747121] * Acc@1 65.690 Acc@5 88.278 loss 1.400
[15:09:45.747266] Accuracy of the network on the 50000 test images: 65.7%
[15:09:45.747288] Max accuracy: 66.01%
[15:09:45.754200] log_dir: ./output_dir_cml_spikformer
[15:09:47.421147] Epoch: [52]  [   0/5004]  eta: 2:18:58  lr: 0.001036  loss: 2.8597 (2.8597)  time: 1.6663  data: 1.0414  max mem: 17867
[15:21:28.056473] Epoch: [52]  [2000/5004]  eta: 0:17:34  lr: 0.001034  loss: 3.2983 (3.2241)  time: 0.3540  data: 0.0002  max mem: 17867
[15:33:07.464469] Epoch: [52]  [4000/5004]  eta: 0:05:51  lr: 0.001031  loss: 3.3746 (3.2378)  time: 0.3512  data: 0.0002  max mem: 17867
[15:38:58.977332] Epoch: [52]  [5003/5004]  eta: 0:00:00  lr: 0.001030  loss: 3.1368 (3.2376)  time: 0.3482  data: 0.0006  max mem: 17867
[15:38:59.335071] Epoch: [52] Total time: 0:29:13 (0.3504 s / it)
[15:38:59.335765] Averaged stats: lr: 0.001030  loss: 3.1368 (3.2344)
[15:39:00.606281] Test:  [   0/1563]  eta: 0:32:59  loss: 0.3255 (0.3255)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.2665  data: 0.8364  max mem: 17867
[15:40:05.901659] Test:  [ 500/1563]  eta: 0:02:21  loss: 1.0389 (1.1578)  acc1: 75.0000 (70.6400)  acc5: 93.7500 (92.2779)  time: 0.1306  data: 0.0002  max mem: 17867
[15:41:11.003233] Test:  [1000/1563]  eta: 0:01:14  loss: 1.4993 (1.3035)  acc1: 65.6250 (67.9914)  acc5: 87.5000 (89.7415)  time: 0.1299  data: 0.0002  max mem: 17867
[15:42:16.022618] Test:  [1500/1563]  eta: 0:00:08  loss: 0.7861 (1.4044)  acc1: 81.2500 (65.7895)  acc5: 93.7500 (88.2058)  time: 0.1299  data: 0.0002  max mem: 17867
[15:42:24.004474] Test:  [1562/1563]  eta: 0:00:00  loss: 0.7546 (1.4057)  acc1: 87.5000 (65.7880)  acc5: 96.8750 (88.1520)  time: 0.1262  data: 0.0001  max mem: 17867
[15:42:24.071931] Test: Total time: 0:03:24 (0.1310 s / it)
[15:42:24.093286] * Acc@1 65.788 Acc@5 88.152 loss 1.406
[15:42:24.093404] Accuracy of the network on the 50000 test images: 65.8%
[15:42:24.093427] Max accuracy: 66.01%
[15:42:24.114101] log_dir: ./output_dir_cml_spikformer
[15:42:25.573611] Epoch: [53]  [   0/5004]  eta: 2:01:35  lr: 0.001030  loss: 2.6714 (2.6714)  time: 1.4579  data: 1.1285  max mem: 17867
[15:54:05.290608] Epoch: [53]  [2000/5004]  eta: 0:17:32  lr: 0.001027  loss: 3.1247 (3.2106)  time: 0.3496  data: 0.0002  max mem: 17867
[16:05:43.420452] Epoch: [53]  [4000/5004]  eta: 0:05:51  lr: 0.001024  loss: 3.1234 (3.2124)  time: 0.3501  data: 0.0002  max mem: 17867
[16:11:33.693186] Epoch: [53]  [5003/5004]  eta: 0:00:00  lr: 0.001023  loss: 3.0276 (3.2161)  time: 0.3456  data: 0.0011  max mem: 17867
[16:11:34.071631] Epoch: [53] Total time: 0:29:09 (0.3497 s / it)
[16:11:34.072352] Averaged stats: lr: 0.001023  loss: 3.0276 (3.2269)
[16:11:35.170438] Test:  [   0/1563]  eta: 0:28:27  loss: 0.3665 (0.3665)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.0927  data: 0.9530  max mem: 17867
[16:12:40.177591] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.4401 (1.1590)  acc1: 62.5000 (71.2762)  acc5: 87.5000 (92.5711)  time: 0.1298  data: 0.0002  max mem: 17867
[16:13:45.127674] Test:  [1000/1563]  eta: 0:01:13  loss: 1.3632 (1.3133)  acc1: 68.7500 (68.4722)  acc5: 90.6250 (90.0755)  time: 0.1298  data: 0.0002  max mem: 17867
[16:14:50.106944] Test:  [1500/1563]  eta: 0:00:08  loss: 0.7791 (1.4111)  acc1: 84.3750 (66.5327)  acc5: 93.7500 (88.5784)  time: 0.1298  data: 0.0002  max mem: 17867
[16:14:58.091461] Test:  [1562/1563]  eta: 0:00:00  loss: 0.8180 (1.4108)  acc1: 87.5000 (66.5180)  acc5: 96.8750 (88.5920)  time: 0.1261  data: 0.0001  max mem: 17867
[16:14:58.160455] Test: Total time: 0:03:24 (0.1306 s / it)
[16:14:58.339684] * Acc@1 66.518 Acc@5 88.592 loss 1.411
[16:14:58.339931] Accuracy of the network on the 50000 test images: 66.5%
[16:14:58.339954] Max accuracy: 66.52%
[16:14:58.368252] log_dir: ./output_dir_cml_spikformer
[16:14:59.811110] Epoch: [54]  [   0/5004]  eta: 2:00:15  lr: 0.001023  loss: 3.2121 (3.2121)  time: 1.4419  data: 1.0438  max mem: 17867
[16:26:37.838658] Epoch: [54]  [2000/5004]  eta: 0:17:30  lr: 0.001020  loss: 3.1836 (3.2259)  time: 0.3486  data: 0.0002  max mem: 17867
[16:38:15.400284] Epoch: [54]  [4000/5004]  eta: 0:05:50  lr: 0.001017  loss: 3.2954 (3.2226)  time: 0.3450  data: 0.0002  max mem: 17867
[16:44:05.066915] Epoch: [54]  [5003/5004]  eta: 0:00:00  lr: 0.001016  loss: 3.2596 (3.2200)  time: 0.3500  data: 0.0012  max mem: 17867
[16:44:05.403518] Epoch: [54] Total time: 0:29:07 (0.3491 s / it)
[16:44:05.414483] Averaged stats: lr: 0.001016  loss: 3.2596 (3.2224)
[16:44:06.604281] Test:  [   0/1563]  eta: 0:30:54  loss: 0.5518 (0.5518)  acc1: 90.6250 (90.6250)  acc5: 96.8750 (96.8750)  time: 1.1862  data: 1.0458  max mem: 17867
[16:45:11.750226] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.1172 (1.1887)  acc1: 68.7500 (69.2802)  acc5: 90.6250 (92.4276)  time: 0.1298  data: 0.0002  max mem: 17867
[16:46:16.707272] Test:  [1000/1563]  eta: 0:01:13  loss: 1.6000 (1.3254)  acc1: 53.1250 (67.1672)  acc5: 87.5000 (89.9382)  time: 0.1299  data: 0.0002  max mem: 17867
[16:47:21.671692] Test:  [1500/1563]  eta: 0:00:08  loss: 0.7311 (1.4196)  acc1: 87.5000 (65.4126)  acc5: 93.7500 (88.3453)  time: 0.1299  data: 0.0002  max mem: 17867
[16:47:29.677561] Test:  [1562/1563]  eta: 0:00:00  loss: 0.7840 (1.4199)  acc1: 87.5000 (65.4220)  acc5: 93.7500 (88.3560)  time: 0.1264  data: 0.0001  max mem: 17867
[16:47:29.740394] Test: Total time: 0:03:24 (0.1307 s / it)
[16:47:30.034064] * Acc@1 65.422 Acc@5 88.356 loss 1.420
[16:47:30.034227] Accuracy of the network on the 50000 test images: 65.4%
[16:47:30.034249] Max accuracy: 66.52%
[16:47:30.080007] log_dir: ./output_dir_cml_spikformer
[16:47:31.510325] Epoch: [55]  [   0/5004]  eta: 1:59:13  lr: 0.001016  loss: 3.2369 (3.2369)  time: 1.4296  data: 1.0841  max mem: 17867
[16:59:10.789131] Epoch: [55]  [2000/5004]  eta: 0:17:31  lr: 0.001013  loss: 3.1974 (3.1910)  time: 0.3498  data: 0.0003  max mem: 17867
[17:10:49.989456] Epoch: [55]  [4000/5004]  eta: 0:05:51  lr: 0.001010  loss: 3.3118 (3.2112)  time: 0.3480  data: 0.0002  max mem: 17867
[17:16:46.098724] Epoch: [55]  [5003/5004]  eta: 0:00:00  lr: 0.001009  loss: 3.4133 (3.2121)  time: 0.3452  data: 0.0006  max mem: 17867
[17:16:46.440738] Epoch: [55] Total time: 0:29:16 (0.3510 s / it)
[17:16:46.441509] Averaged stats: lr: 0.001009  loss: 3.4133 (3.2187)
[17:16:47.454459] Test:  [   0/1563]  eta: 0:26:17  loss: 0.4775 (0.4775)  acc1: 90.6250 (90.6250)  acc5: 96.8750 (96.8750)  time: 1.0093  data: 0.8586  max mem: 17867
[17:17:52.440438] Test:  [ 500/1563]  eta: 0:02:19  loss: 1.2828 (1.1043)  acc1: 68.7500 (71.9311)  acc5: 93.7500 (93.1387)  time: 0.1299  data: 0.0002  max mem: 17867
[17:18:57.402871] Test:  [1000/1563]  eta: 0:01:13  loss: 1.9022 (1.2711)  acc1: 53.1250 (68.8561)  acc5: 81.2500 (90.3066)  time: 0.1301  data: 0.0002  max mem: 17867
[17:20:02.422732] Test:  [1500/1563]  eta: 0:00:08  loss: 0.7360 (1.3631)  acc1: 84.3750 (66.8346)  acc5: 93.7500 (88.8553)  time: 0.1299  data: 0.0002  max mem: 17867
[17:20:10.726636] Test:  [1562/1563]  eta: 0:00:00  loss: 0.7311 (1.3633)  acc1: 84.3750 (66.8340)  acc5: 93.7500 (88.8400)  time: 0.1422  data: 0.0001  max mem: 17867
[17:20:10.878524] Test: Total time: 0:03:24 (0.1308 s / it)
[17:20:10.880080] * Acc@1 66.834 Acc@5 88.840 loss 1.363
[17:20:10.880195] Accuracy of the network on the 50000 test images: 66.8%
[17:20:10.880216] Max accuracy: 66.83%
[17:20:10.906121] log_dir: ./output_dir_cml_spikformer
[17:20:12.300163] Epoch: [56]  [   0/5004]  eta: 1:56:11  lr: 0.001009  loss: 3.1354 (3.1354)  time: 1.3931  data: 0.8765  max mem: 17867
[17:31:52.834255] Epoch: [56]  [2000/5004]  eta: 0:17:33  lr: 0.001006  loss: 3.1158 (3.1987)  time: 0.3530  data: 0.0002  max mem: 17867
[17:43:31.813597] Epoch: [56]  [4000/5004]  eta: 0:05:51  lr: 0.001003  loss: 3.2447 (3.2108)  time: 0.3457  data: 0.0002  max mem: 17867
[17:49:22.873667] Epoch: [56]  [5003/5004]  eta: 0:00:00  lr: 0.001002  loss: 3.1776 (3.2166)  time: 0.3464  data: 0.0011  max mem: 17867
[17:49:23.225746] Epoch: [56] Total time: 0:29:12 (0.3502 s / it)
[17:49:23.226535] Averaged stats: lr: 0.001002  loss: 3.1776 (3.2117)
[17:49:24.201684] Test:  [   0/1563]  eta: 0:25:18  loss: 0.2735 (0.2735)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 0.9715  data: 0.8170  max mem: 17867
[17:50:29.197717] Test:  [ 500/1563]  eta: 0:02:19  loss: 1.2368 (1.1328)  acc1: 68.7500 (71.2076)  acc5: 93.7500 (92.5711)  time: 0.1302  data: 0.0002  max mem: 17867
[17:51:34.384969] Test:  [1000/1563]  eta: 0:01:13  loss: 1.8038 (1.2858)  acc1: 56.2500 (68.3192)  acc5: 84.3750 (90.2223)  time: 0.1299  data: 0.0002  max mem: 17867
[17:52:39.371457] Test:  [1500/1563]  eta: 0:00:08  loss: 0.7000 (1.3826)  acc1: 84.3750 (66.5452)  acc5: 96.8750 (88.7991)  time: 0.1301  data: 0.0002  max mem: 17867
[17:52:47.358181] Test:  [1562/1563]  eta: 0:00:00  loss: 0.6005 (1.3816)  acc1: 87.5000 (66.5720)  acc5: 96.8750 (88.7900)  time: 0.1263  data: 0.0001  max mem: 17867
[17:52:47.449432] Test: Total time: 0:03:24 (0.1307 s / it)
[17:52:47.602917] * Acc@1 66.572 Acc@5 88.790 loss 1.382
[17:52:47.603057] Accuracy of the network on the 50000 test images: 66.6%
[17:52:47.603080] Max accuracy: 66.83%
[17:52:47.639370] log_dir: ./output_dir_cml_spikformer
[17:52:49.083091] Epoch: [57]  [   0/5004]  eta: 2:00:17  lr: 0.001002  loss: 2.9556 (2.9556)  time: 1.4424  data: 0.9107  max mem: 17867
[18:04:29.218122] Epoch: [57]  [2000/5004]  eta: 0:17:33  lr: 0.000999  loss: 3.0436 (3.2054)  time: 0.3483  data: 0.0002  max mem: 17867
[18:16:08.712315] Epoch: [57]  [4000/5004]  eta: 0:05:51  lr: 0.000996  loss: 3.2437 (3.2067)  time: 0.3521  data: 0.0002  max mem: 17867
[18:21:59.324190] Epoch: [57]  [5003/5004]  eta: 0:00:00  lr: 0.000994  loss: 3.2427 (3.2095)  time: 0.3488  data: 0.0012  max mem: 17867
[18:21:59.703643] Epoch: [57] Total time: 0:29:12 (0.3501 s / it)
[18:21:59.704496] Averaged stats: lr: 0.000994  loss: 3.2427 (3.2082)
[18:22:00.774001] Test:  [   0/1563]  eta: 0:27:44  loss: 0.3461 (0.3461)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.0648  data: 0.9118  max mem: 17867
[18:23:05.780193] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.1956 (1.1627)  acc1: 75.0000 (71.2388)  acc5: 93.7500 (92.5274)  time: 0.1299  data: 0.0002  max mem: 17867
[18:24:10.772907] Test:  [1000/1563]  eta: 0:01:13  loss: 1.6500 (1.3234)  acc1: 56.2500 (68.1256)  acc5: 87.5000 (89.7384)  time: 0.1299  data: 0.0002  max mem: 17867
[18:25:15.762989] Test:  [1500/1563]  eta: 0:00:08  loss: 0.8796 (1.4148)  acc1: 78.1250 (66.2017)  acc5: 93.7500 (88.3494)  time: 0.1299  data: 0.0002  max mem: 17867
[18:25:23.749469] Test:  [1562/1563]  eta: 0:00:00  loss: 0.6744 (1.4137)  acc1: 81.2500 (66.1580)  acc5: 93.7500 (88.3940)  time: 0.1264  data: 0.0001  max mem: 17867
[18:25:23.828281] Test: Total time: 0:03:24 (0.1306 s / it)
[18:25:23.989636] * Acc@1 66.158 Acc@5 88.394 loss 1.414
[18:25:23.989814] Accuracy of the network on the 50000 test images: 66.2%
[18:25:23.989842] Max accuracy: 66.83%
[18:25:23.996576] log_dir: ./output_dir_cml_spikformer
[18:25:25.646546] Epoch: [58]  [   0/5004]  eta: 2:17:30  lr: 0.000994  loss: 2.7023 (2.7023)  time: 1.6488  data: 1.0059  max mem: 17867
[18:37:06.190057] Epoch: [58]  [2000/5004]  eta: 0:17:34  lr: 0.000991  loss: 3.1248 (3.2020)  time: 0.3487  data: 0.0003  max mem: 17867
[18:48:46.263981] Epoch: [58]  [4000/5004]  eta: 0:05:51  lr: 0.000989  loss: 3.0894 (3.2062)  time: 0.3468  data: 0.0002  max mem: 17867
[18:54:37.227187] Epoch: [58]  [5003/5004]  eta: 0:00:00  lr: 0.000987  loss: 3.1831 (3.2054)  time: 0.3463  data: 0.0011  max mem: 17867
[18:54:37.589174] Epoch: [58] Total time: 0:29:13 (0.3504 s / it)
[18:54:37.589946] Averaged stats: lr: 0.000987  loss: 3.1831 (3.2048)
[18:54:39.004324] Test:  [   0/1563]  eta: 0:36:44  loss: 0.3429 (0.3429)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.4107  data: 1.2732  max mem: 17867
[18:55:43.970111] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.1814 (1.1533)  acc1: 75.0000 (70.3905)  acc5: 90.6250 (92.3278)  time: 0.1299  data: 0.0002  max mem: 17867
[18:56:48.940279] Test:  [1000/1563]  eta: 0:01:13  loss: 1.6917 (1.3204)  acc1: 56.2500 (67.7229)  acc5: 84.3750 (89.5042)  time: 0.1298  data: 0.0002  max mem: 17867
[18:57:53.912650] Test:  [1500/1563]  eta: 0:00:08  loss: 0.6686 (1.4133)  acc1: 84.3750 (65.6021)  acc5: 96.8750 (88.1225)  time: 0.1299  data: 0.0002  max mem: 17867
[18:58:01.890585] Test:  [1562/1563]  eta: 0:00:00  loss: 0.6112 (1.4139)  acc1: 81.2500 (65.5400)  acc5: 93.7500 (88.1060)  time: 0.1261  data: 0.0001  max mem: 17867
[18:58:01.964309] Test: Total time: 0:03:24 (0.1308 s / it)
[18:58:02.237981] * Acc@1 65.540 Acc@5 88.106 loss 1.414
[18:58:02.238178] Accuracy of the network on the 50000 test images: 65.5%
[18:58:02.238203] Max accuracy: 66.83%
[18:58:02.271567] log_dir: ./output_dir_cml_spikformer
[18:58:03.857195] Epoch: [59]  [   0/5004]  eta: 2:12:11  lr: 0.000987  loss: 3.6023 (3.6023)  time: 1.5850  data: 1.0695  max mem: 17867
[19:09:45.371915] Epoch: [59]  [2000/5004]  eta: 0:17:35  lr: 0.000984  loss: 3.1187 (3.1864)  time: 0.3491  data: 0.0004  max mem: 17867
[19:21:26.146208] Epoch: [59]  [4000/5004]  eta: 0:05:52  lr: 0.000981  loss: 3.0492 (3.1961)  time: 0.3538  data: 0.0002  max mem: 17867
[19:27:17.369463] Epoch: [59]  [5003/5004]  eta: 0:00:00  lr: 0.000980  loss: 3.2227 (3.1979)  time: 0.3434  data: 0.0011  max mem: 17867
[19:27:17.727026] Epoch: [59] Total time: 0:29:15 (0.3508 s / it)
[19:27:17.728873] Averaged stats: lr: 0.000980  loss: 3.2227 (3.1973)
[19:27:18.765710] Test:  [   0/1563]  eta: 0:26:55  loss: 0.9843 (0.9843)  acc1: 84.3750 (84.3750)  acc5: 93.7500 (93.7500)  time: 1.0334  data: 0.8691  max mem: 17867
[19:28:23.836694] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.0347 (1.0759)  acc1: 71.8750 (72.3241)  acc5: 93.7500 (93.2822)  time: 0.1300  data: 0.0002  max mem: 17867
[19:29:28.807351] Test:  [1000/1563]  eta: 0:01:13  loss: 1.5137 (1.2510)  acc1: 59.3750 (69.1777)  acc5: 87.5000 (90.5626)  time: 0.1299  data: 0.0002  max mem: 17867
[19:30:33.798078] Test:  [1500/1563]  eta: 0:00:08  loss: 0.7120 (1.3572)  acc1: 81.2500 (66.9262)  acc5: 93.7500 (88.9220)  time: 0.1299  data: 0.0002  max mem: 17867
[19:30:41.778468] Test:  [1562/1563]  eta: 0:00:00  loss: 0.6689 (1.3543)  acc1: 84.3750 (66.9840)  acc5: 93.7500 (88.9680)  time: 0.1262  data: 0.0001  max mem: 17867
[19:30:41.834485] Test: Total time: 0:03:24 (0.1306 s / it)
[19:30:42.039726] * Acc@1 66.984 Acc@5 88.968 loss 1.354
[19:30:42.039870] Accuracy of the network on the 50000 test images: 67.0%
[19:30:42.039894] Max accuracy: 66.98%
[19:30:42.066318] log_dir: ./output_dir_cml_spikformer
[19:30:43.615959] Epoch: [60]  [   0/5004]  eta: 2:09:09  lr: 0.000980  loss: 3.2620 (3.2620)  time: 1.5486  data: 1.0719  max mem: 17867
[19:42:24.927715] Epoch: [60]  [2000/5004]  eta: 0:17:35  lr: 0.000977  loss: 3.2298 (3.1826)  time: 0.3529  data: 0.0002  max mem: 17867
[19:54:05.063694] Epoch: [60]  [4000/5004]  eta: 0:05:52  lr: 0.000974  loss: 3.1353 (3.1879)  time: 0.3463  data: 0.0002  max mem: 17867
[19:59:55.895903] Epoch: [60]  [5003/5004]  eta: 0:00:00  lr: 0.000972  loss: 3.1664 (3.1899)  time: 0.3447  data: 0.0011  max mem: 17867
[19:59:56.274429] Epoch: [60] Total time: 0:29:14 (0.3506 s / it)
[19:59:56.275130] Averaged stats: lr: 0.000972  loss: 3.1664 (3.1926)
[19:59:57.371341] Test:  [   0/1563]  eta: 0:28:27  loss: 0.4052 (0.4052)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.0926  data: 0.9587  max mem: 17867
[20:01:02.456495] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.0702 (1.1095)  acc1: 71.8750 (71.8313)  acc5: 93.7500 (92.8081)  time: 0.1304  data: 0.0002  max mem: 17867
[20:02:07.473702] Test:  [1000/1563]  eta: 0:01:13  loss: 1.5009 (1.2687)  acc1: 59.3750 (68.6876)  acc5: 90.6250 (90.3034)  time: 0.1310  data: 0.0002  max mem: 17867
[20:03:12.620575] Test:  [1500/1563]  eta: 0:00:08  loss: 0.7377 (1.3633)  acc1: 84.3750 (66.7305)  acc5: 93.7500 (88.7825)  time: 0.1300  data: 0.0002  max mem: 17867
[20:03:20.606724] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4631 (1.3600)  acc1: 87.5000 (66.7760)  acc5: 96.8750 (88.8640)  time: 0.1262  data: 0.0001  max mem: 17867
[20:03:20.699950] Test: Total time: 0:03:24 (0.1308 s / it)
[20:03:20.829111] * Acc@1 66.776 Acc@5 88.864 loss 1.360
[20:03:20.829301] Accuracy of the network on the 50000 test images: 66.8%
[20:03:20.829325] Max accuracy: 66.98%
[20:03:20.843748] log_dir: ./output_dir_cml_spikformer
[20:03:22.397662] Epoch: [61]  [   0/5004]  eta: 2:09:32  lr: 0.000972  loss: 3.4757 (3.4757)  time: 1.5532  data: 0.9974  max mem: 17867
[20:15:02.315063] Epoch: [61]  [2000/5004]  eta: 0:17:33  lr: 0.000969  loss: 3.0723 (3.2016)  time: 0.3480  data: 0.0002  max mem: 17867
[20:26:42.212028] Epoch: [61]  [4000/5004]  eta: 0:05:51  lr: 0.000966  loss: 3.0753 (3.1942)  time: 0.3535  data: 0.0002  max mem: 17867
[20:32:33.241391] Epoch: [61]  [5003/5004]  eta: 0:00:00  lr: 0.000964  loss: 3.2189 (3.1917)  time: 0.3452  data: 0.0006  max mem: 17867
[20:32:33.568251] Epoch: [61] Total time: 0:29:12 (0.3503 s / it)
[20:32:33.579284] Averaged stats: lr: 0.000964  loss: 3.2189 (3.1893)
[20:32:34.635300] Test:  [   0/1563]  eta: 0:27:24  loss: 0.5930 (0.5930)  acc1: 90.6250 (90.6250)  acc5: 93.7500 (93.7500)  time: 1.0523  data: 0.9122  max mem: 17867
[20:33:39.592456] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.2270 (1.0730)  acc1: 62.5000 (72.7483)  acc5: 90.6250 (93.2947)  time: 0.1298  data: 0.0002  max mem: 17867
[20:34:44.555811] Test:  [1000/1563]  eta: 0:01:13  loss: 1.6248 (1.2709)  acc1: 59.3750 (68.9873)  acc5: 87.5000 (90.2004)  time: 0.1298  data: 0.0002  max mem: 17867
[20:35:49.548405] Test:  [1500/1563]  eta: 0:00:08  loss: 0.8063 (1.3805)  acc1: 81.2500 (66.7263)  acc5: 93.7500 (88.6909)  time: 0.1298  data: 0.0002  max mem: 17867
[20:35:57.526585] Test:  [1562/1563]  eta: 0:00:00  loss: 0.6268 (1.3796)  acc1: 87.5000 (66.7380)  acc5: 96.8750 (88.7200)  time: 0.1262  data: 0.0001  max mem: 17867
[20:35:57.608710] Test: Total time: 0:03:24 (0.1305 s / it)
[20:35:57.834226] * Acc@1 66.738 Acc@5 88.720 loss 1.380
[20:35:57.834362] Accuracy of the network on the 50000 test images: 66.7%
[20:35:57.834384] Max accuracy: 66.98%
[20:35:57.845750] log_dir: ./output_dir_cml_spikformer
[20:35:59.314606] Epoch: [62]  [   0/5004]  eta: 2:02:26  lr: 0.000964  loss: 2.9517 (2.9517)  time: 1.4681  data: 0.9498  max mem: 17867
[20:47:38.539352] Epoch: [62]  [2000/5004]  eta: 0:17:31  lr: 0.000961  loss: 3.0585 (3.1804)  time: 0.3499  data: 0.0002  max mem: 17867
[20:59:19.131950] Epoch: [62]  [4000/5004]  eta: 0:05:51  lr: 0.000958  loss: 3.2659 (3.1850)  time: 0.3465  data: 0.0002  max mem: 17867
[21:05:09.838047] Epoch: [62]  [5003/5004]  eta: 0:00:00  lr: 0.000957  loss: 3.2144 (3.1879)  time: 0.3440  data: 0.0007  max mem: 17867
[21:05:10.176857] Epoch: [62] Total time: 0:29:12 (0.3502 s / it)
[21:05:10.181214] Averaged stats: lr: 0.000957  loss: 3.2144 (3.1835)
[21:05:11.502820] Test:  [   0/1563]  eta: 0:34:19  loss: 0.3315 (0.3315)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.3178  data: 1.1798  max mem: 17867
[21:06:16.554716] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.9911 (1.0961)  acc1: 68.7500 (72.3553)  acc5: 90.6250 (92.7582)  time: 0.1299  data: 0.0002  max mem: 17867
[21:07:21.517063] Test:  [1000/1563]  eta: 0:01:13  loss: 1.6474 (1.2568)  acc1: 65.6250 (69.0778)  acc5: 90.6250 (90.2878)  time: 0.1299  data: 0.0002  max mem: 17867
[21:08:26.474576] Test:  [1500/1563]  eta: 0:00:08  loss: 0.7996 (1.3285)  acc1: 78.1250 (67.6882)  acc5: 93.7500 (89.0781)  time: 0.1299  data: 0.0002  max mem: 17867
[21:08:34.456944] Test:  [1562/1563]  eta: 0:00:00  loss: 0.6643 (1.3263)  acc1: 84.3750 (67.7160)  acc5: 96.8750 (89.1120)  time: 0.1261  data: 0.0001  max mem: 17867
[21:08:34.525938] Test: Total time: 0:03:24 (0.1307 s / it)
[21:08:34.735997] * Acc@1 67.716 Acc@5 89.112 loss 1.326
[21:08:34.736275] Accuracy of the network on the 50000 test images: 67.7%
[21:08:34.736299] Max accuracy: 67.72%
[21:08:34.769320] log_dir: ./output_dir_cml_spikformer
[21:08:36.215911] Epoch: [63]  [   0/5004]  eta: 2:00:34  lr: 0.000957  loss: 3.2075 (3.2075)  time: 1.4457  data: 1.0361  max mem: 17867
[21:20:16.039812] Epoch: [63]  [2000/5004]  eta: 0:17:32  lr: 0.000954  loss: 3.1738 (3.1743)  time: 0.3486  data: 0.0002  max mem: 17867
[21:31:56.046085] Epoch: [63]  [4000/5004]  eta: 0:05:51  lr: 0.000951  loss: 3.2483 (3.1707)  time: 0.3509  data: 0.0002  max mem: 17867
[21:37:47.712160] Epoch: [63]  [5003/5004]  eta: 0:00:00  lr: 0.000949  loss: 3.1925 (3.1751)  time: 0.3454  data: 0.0013  max mem: 17867
[21:37:48.077382] Epoch: [63] Total time: 0:29:13 (0.3504 s / it)
[21:37:48.078286] Averaged stats: lr: 0.000949  loss: 3.1925 (3.1782)
[21:37:49.143963] Test:  [   0/1563]  eta: 0:27:37  loss: 0.5217 (0.5217)  acc1: 90.6250 (90.6250)  acc5: 96.8750 (96.8750)  time: 1.0608  data: 0.8893  max mem: 17867
[21:38:54.157959] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.9798 (1.0745)  acc1: 75.0000 (72.3802)  acc5: 93.7500 (92.8643)  time: 0.1299  data: 0.0002  max mem: 17867
[21:39:59.152311] Test:  [1000/1563]  eta: 0:01:13  loss: 1.3633 (1.2358)  acc1: 62.5000 (69.5617)  acc5: 90.6250 (90.3721)  time: 0.1299  data: 0.0002  max mem: 17867
[21:41:04.206881] Test:  [1500/1563]  eta: 0:00:08  loss: 0.7473 (1.3382)  acc1: 84.3750 (67.4800)  acc5: 93.7500 (88.8637)  time: 0.1300  data: 0.0002  max mem: 17867
[21:41:12.206280] Test:  [1562/1563]  eta: 0:00:00  loss: 0.6213 (1.3352)  acc1: 87.5000 (67.5460)  acc5: 96.8750 (88.9380)  time: 0.1262  data: 0.0001  max mem: 17867
[21:41:12.284616] Test: Total time: 0:03:24 (0.1306 s / it)
[21:41:12.528390] * Acc@1 67.546 Acc@5 88.938 loss 1.335
[21:41:12.528532] Accuracy of the network on the 50000 test images: 67.5%
[21:41:12.528554] Max accuracy: 67.72%
[21:41:12.566839] log_dir: ./output_dir_cml_spikformer
[21:41:14.456600] Epoch: [64]  [   0/5004]  eta: 2:37:32  lr: 0.000949  loss: 2.9430 (2.9430)  time: 1.8889  data: 0.9717  max mem: 17867
[21:52:54.478071] Epoch: [64]  [2000/5004]  eta: 0:17:33  lr: 0.000946  loss: 3.3349 (3.1653)  time: 0.3575  data: 0.0002  max mem: 17867
[22:04:33.274258] Epoch: [64]  [4000/5004]  eta: 0:05:51  lr: 0.000943  loss: 3.1590 (3.1710)  time: 0.3517  data: 0.0002  max mem: 17867
[22:10:24.215924] Epoch: [64]  [5003/5004]  eta: 0:00:00  lr: 0.000941  loss: 3.0408 (3.1714)  time: 0.3447  data: 0.0006  max mem: 17867
[22:10:24.560571] Epoch: [64] Total time: 0:29:11 (0.3501 s / it)
[22:10:24.578063] Averaged stats: lr: 0.000941  loss: 3.0408 (3.1726)
[22:10:25.782903] Test:  [   0/1563]  eta: 0:31:17  loss: 0.4314 (0.4314)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.2012  data: 1.0595  max mem: 17867
[22:11:30.902761] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.1752 (1.0729)  acc1: 68.7500 (72.4800)  acc5: 93.7500 (93.4444)  time: 0.1299  data: 0.0002  max mem: 17867
[22:12:35.881648] Test:  [1000/1563]  eta: 0:01:13  loss: 1.5006 (1.2418)  acc1: 65.6250 (69.3088)  acc5: 90.6250 (90.7530)  time: 0.1299  data: 0.0002  max mem: 17867
[22:13:40.872459] Test:  [1500/1563]  eta: 0:00:08  loss: 0.7210 (1.3382)  acc1: 84.3750 (67.5237)  acc5: 93.7500 (89.3279)  time: 0.1299  data: 0.0002  max mem: 17867
[22:13:48.855540] Test:  [1562/1563]  eta: 0:00:00  loss: 0.6268 (1.3364)  acc1: 81.2500 (67.5720)  acc5: 96.8750 (89.3600)  time: 0.1262  data: 0.0001  max mem: 17867
[22:13:48.927940] Test: Total time: 0:03:24 (0.1307 s / it)
[22:13:49.198893] * Acc@1 67.572 Acc@5 89.360 loss 1.336
[22:13:49.199114] Accuracy of the network on the 50000 test images: 67.6%
[22:13:49.199137] Max accuracy: 67.72%
[22:13:49.247883] log_dir: ./output_dir_cml_spikformer
[22:13:50.753511] Epoch: [65]  [   0/5004]  eta: 2:05:28  lr: 0.000941  loss: 2.8250 (2.8250)  time: 1.5046  data: 1.0612  max mem: 17867
[22:25:31.364085] Epoch: [65]  [2000/5004]  eta: 0:17:34  lr: 0.000938  loss: 3.1507 (3.1706)  time: 0.3521  data: 0.0002  max mem: 17867
[22:37:11.020107] Epoch: [65]  [4000/5004]  eta: 0:05:51  lr: 0.000935  loss: 3.0519 (3.1690)  time: 0.3444  data: 0.0003  max mem: 17867
[22:43:01.926464] Epoch: [65]  [5003/5004]  eta: 0:00:00  lr: 0.000933  loss: 3.2022 (3.1747)  time: 0.3450  data: 0.0011  max mem: 17867
[22:43:02.301074] Epoch: [65] Total time: 0:29:13 (0.3503 s / it)
[22:43:02.302354] Averaged stats: lr: 0.000933  loss: 3.2022 (3.1679)
[22:43:03.448665] Test:  [   0/1563]  eta: 0:29:46  loss: 0.4077 (0.4077)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.1428  data: 1.0056  max mem: 17867
[22:44:08.555992] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.1809 (1.0688)  acc1: 68.7500 (72.4551)  acc5: 90.6250 (93.1761)  time: 0.1301  data: 0.0002  max mem: 17867
[22:45:13.557206] Test:  [1000/1563]  eta: 0:01:13  loss: 1.4559 (1.2292)  acc1: 59.3750 (69.7428)  acc5: 87.5000 (90.5938)  time: 0.1300  data: 0.0002  max mem: 17867
[22:46:18.656965] Test:  [1500/1563]  eta: 0:00:08  loss: 0.9025 (1.3352)  acc1: 78.1250 (67.5654)  acc5: 93.7500 (89.1489)  time: 0.1299  data: 0.0002  max mem: 17867
[22:46:26.644882] Test:  [1562/1563]  eta: 0:00:00  loss: 0.5114 (1.3329)  acc1: 87.5000 (67.6140)  acc5: 96.8750 (89.1700)  time: 0.1263  data: 0.0001  max mem: 17867
[22:46:26.714853] Test: Total time: 0:03:24 (0.1308 s / it)
[22:46:26.900604] * Acc@1 67.614 Acc@5 89.170 loss 1.333
[22:46:26.900809] Accuracy of the network on the 50000 test images: 67.6%
[22:46:26.900839] Max accuracy: 67.72%
[22:46:26.940475] log_dir: ./output_dir_cml_spikformer
[22:46:28.724987] Epoch: [66]  [   0/5004]  eta: 2:28:45  lr: 0.000933  loss: 3.6975 (3.6975)  time: 1.7836  data: 1.0548  max mem: 17867
[22:58:09.081142] Epoch: [66]  [2000/5004]  eta: 0:17:34  lr: 0.000930  loss: 3.1225 (3.1518)  time: 0.3508  data: 0.0002  max mem: 17867
[23:09:49.628411] Epoch: [66]  [4000/5004]  eta: 0:05:51  lr: 0.000927  loss: 3.1183 (3.1627)  time: 0.3527  data: 0.0002  max mem: 17867
[23:15:40.702340] Epoch: [66]  [5003/5004]  eta: 0:00:00  lr: 0.000925  loss: 3.0541 (3.1660)  time: 0.3455  data: 0.0012  max mem: 17867
[23:15:41.054454] Epoch: [66] Total time: 0:29:14 (0.3505 s / it)
[23:15:41.055179] Averaged stats: lr: 0.000925  loss: 3.0541 (3.1602)
[23:15:45.375693] Test:  [   0/1563]  eta: 1:52:27  loss: 0.5109 (0.5109)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 4.3169  data: 4.1689  max mem: 17867
[23:16:50.476556] Test:  [ 500/1563]  eta: 0:02:27  loss: 1.0998 (1.1032)  acc1: 71.8750 (71.4633)  acc5: 96.8750 (93.0015)  time: 0.1299  data: 0.0002  max mem: 17867
[23:17:55.455244] Test:  [1000/1563]  eta: 0:01:15  loss: 1.5241 (1.2464)  acc1: 56.2500 (68.7968)  acc5: 90.6250 (90.6531)  time: 0.1299  data: 0.0002  max mem: 17867
[23:19:00.565530] Test:  [1500/1563]  eta: 0:00:08  loss: 0.7335 (1.3304)  acc1: 84.3750 (67.3322)  acc5: 96.8750 (89.3113)  time: 0.1299  data: 0.0002  max mem: 17867
[23:19:08.549002] Test:  [1562/1563]  eta: 0:00:00  loss: 0.6776 (1.3329)  acc1: 87.5000 (67.3500)  acc5: 93.7500 (89.2880)  time: 0.1263  data: 0.0001  max mem: 17867
[23:19:08.612253] Test: Total time: 0:03:27 (0.1328 s / it)
[23:19:08.814414] * Acc@1 67.350 Acc@5 89.288 loss 1.333
[23:19:08.814561] Accuracy of the network on the 50000 test images: 67.3%
[23:19:08.814583] Max accuracy: 67.72%
[23:19:08.837040] log_dir: ./output_dir_cml_spikformer
[23:19:10.484273] Epoch: [67]  [   0/5004]  eta: 2:17:19  lr: 0.000925  loss: 2.5791 (2.5791)  time: 1.6466  data: 1.0729  max mem: 17867
[23:30:50.459764] Epoch: [67]  [2000/5004]  eta: 0:17:33  lr: 0.000922  loss: 3.1261 (3.1507)  time: 0.3494  data: 0.0002  max mem: 17867
[23:42:29.131138] Epoch: [67]  [4000/5004]  eta: 0:05:51  lr: 0.000918  loss: 3.0572 (3.1604)  time: 0.3506  data: 0.0002  max mem: 17867
[23:48:19.185138] Epoch: [67]  [5003/5004]  eta: 0:00:00  lr: 0.000917  loss: 3.2001 (3.1652)  time: 0.3470  data: 0.0006  max mem: 17867
[23:48:19.547937] Epoch: [67] Total time: 0:29:10 (0.3499 s / it)
[23:48:19.551373] Averaged stats: lr: 0.000917  loss: 3.2001 (3.1588)
[23:48:20.569247] Test:  [   0/1563]  eta: 0:26:25  loss: 0.4678 (0.4678)  acc1: 90.6250 (90.6250)  acc5: 96.8750 (96.8750)  time: 1.0144  data: 0.8595  max mem: 17867
[23:49:25.559299] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.2295 (1.1066)  acc1: 68.7500 (71.8313)  acc5: 90.6250 (93.1637)  time: 0.1298  data: 0.0002  max mem: 17867
[23:50:30.494372] Test:  [1000/1563]  eta: 0:01:13  loss: 1.5713 (1.2760)  acc1: 59.3750 (68.8499)  acc5: 90.6250 (90.5407)  time: 0.1298  data: 0.0002  max mem: 17867
[23:51:35.516768] Test:  [1500/1563]  eta: 0:00:08  loss: 0.9520 (1.3739)  acc1: 81.2500 (67.0803)  acc5: 93.7500 (88.9511)  time: 0.1299  data: 0.0002  max mem: 17867
[23:51:43.507033] Test:  [1562/1563]  eta: 0:00:00  loss: 0.5127 (1.3747)  acc1: 87.5000 (67.0720)  acc5: 96.8750 (88.9380)  time: 0.1262  data: 0.0001  max mem: 17867
[23:51:43.569466] Test: Total time: 0:03:24 (0.1305 s / it)
[23:51:43.858039] * Acc@1 67.072 Acc@5 88.938 loss 1.375
[23:51:43.858176] Accuracy of the network on the 50000 test images: 67.1%
[23:51:43.858197] Max accuracy: 67.72%
[23:51:43.884575] log_dir: ./output_dir_cml_spikformer
[23:51:45.624638] Epoch: [68]  [   0/5004]  eta: 2:25:03  lr: 0.000917  loss: 2.8729 (2.8729)  time: 1.7393  data: 1.1625  max mem: 17867
[00:03:25.594754] Epoch: [68]  [2000/5004]  eta: 0:17:33  lr: 0.000914  loss: 3.0325 (3.1359)  time: 0.3521  data: 0.0002  max mem: 17867
[00:15:04.362089] Epoch: [68]  [4000/5004]  eta: 0:05:51  lr: 0.000910  loss: 2.9093 (3.1483)  time: 0.3447  data: 0.0002  max mem: 17867
[00:20:54.355440] Epoch: [68]  [5003/5004]  eta: 0:00:00  lr: 0.000909  loss: 3.0674 (3.1498)  time: 0.3461  data: 0.0011  max mem: 17867
[00:20:54.710315] Epoch: [68] Total time: 0:29:10 (0.3499 s / it)
[00:20:54.717082] Averaged stats: lr: 0.000909  loss: 3.0674 (3.1539)
[00:20:55.779532] Test:  [   0/1563]  eta: 0:27:34  loss: 0.7664 (0.7664)  acc1: 84.3750 (84.3750)  acc5: 93.7500 (93.7500)  time: 1.0588  data: 0.9120  max mem: 17867
[00:22:00.948355] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.1827 (1.0760)  acc1: 65.6250 (71.8875)  acc5: 90.6250 (92.9391)  time: 0.1311  data: 0.0002  max mem: 17867
[00:23:06.282414] Test:  [1000/1563]  eta: 0:01:13  loss: 1.3562 (1.2517)  acc1: 65.6250 (68.8374)  acc5: 90.6250 (90.1754)  time: 0.1300  data: 0.0002  max mem: 17867
[00:24:11.254341] Test:  [1500/1563]  eta: 0:00:08  loss: 0.6756 (1.3380)  acc1: 84.3750 (67.1677)  acc5: 93.7500 (88.9532)  time: 0.1299  data: 0.0002  max mem: 17867
[00:24:19.236597] Test:  [1562/1563]  eta: 0:00:00  loss: 0.6153 (1.3344)  acc1: 84.3750 (67.2380)  acc5: 93.7500 (89.0140)  time: 0.1262  data: 0.0001  max mem: 17867
[00:24:19.295568] Test: Total time: 0:03:24 (0.1309 s / it)
[00:24:19.316759] * Acc@1 67.238 Acc@5 89.014 loss 1.334
[00:24:19.316900] Accuracy of the network on the 50000 test images: 67.2%
[00:24:19.316925] Max accuracy: 67.72%
[00:24:19.323872] log_dir: ./output_dir_cml_spikformer
[00:24:20.925797] Epoch: [69]  [   0/5004]  eta: 2:13:33  lr: 0.000909  loss: 3.4326 (3.4326)  time: 1.6014  data: 1.0928  max mem: 17867
[00:35:59.870568] Epoch: [69]  [2000/5004]  eta: 0:17:31  lr: 0.000905  loss: 3.0021 (3.1419)  time: 0.3535  data: 0.0002  max mem: 17867
[00:47:38.423388] Epoch: [69]  [4000/5004]  eta: 0:05:51  lr: 0.000902  loss: 3.2016 (3.1501)  time: 0.3500  data: 0.0003  max mem: 17867
[00:53:28.453857] Epoch: [69]  [5003/5004]  eta: 0:00:00  lr: 0.000900  loss: 3.1879 (3.1563)  time: 0.3473  data: 0.0011  max mem: 17867
[00:53:28.793468] Epoch: [69] Total time: 0:29:09 (0.3496 s / it)
[00:53:28.800560] Averaged stats: lr: 0.000900  loss: 3.1879 (3.1502)
[00:53:29.863379] Test:  [   0/1563]  eta: 0:27:09  loss: 0.6151 (0.6151)  acc1: 90.6250 (90.6250)  acc5: 96.8750 (96.8750)  time: 1.0427  data: 0.8693  max mem: 17867
[00:54:35.030035] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.1063 (1.1124)  acc1: 65.6250 (72.0497)  acc5: 90.6250 (92.7270)  time: 0.1299  data: 0.0002  max mem: 17867
[00:55:40.185759] Test:  [1000/1563]  eta: 0:01:13  loss: 1.6003 (1.2963)  acc1: 56.2500 (68.4722)  acc5: 87.5000 (90.0724)  time: 0.1300  data: 0.0002  max mem: 17867
[00:56:45.163312] Test:  [1500/1563]  eta: 0:00:08  loss: 0.9792 (1.3936)  acc1: 75.0000 (66.3037)  acc5: 93.7500 (88.6763)  time: 0.1299  data: 0.0002  max mem: 17867
[00:56:53.146263] Test:  [1562/1563]  eta: 0:00:00  loss: 0.6381 (1.3970)  acc1: 87.5000 (66.3240)  acc5: 93.7500 (88.6240)  time: 0.1262  data: 0.0001  max mem: 17867
[00:56:53.210761] Test: Total time: 0:03:24 (0.1308 s / it)
[00:56:53.211916] * Acc@1 66.324 Acc@5 88.624 loss 1.397
[00:56:53.212034] Accuracy of the network on the 50000 test images: 66.3%
[00:56:53.212056] Max accuracy: 67.72%
[00:56:53.240061] log_dir: ./output_dir_cml_spikformer
[00:56:54.722554] Epoch: [70]  [   0/5004]  eta: 2:03:35  lr: 0.000900  loss: 2.7516 (2.7516)  time: 1.4818  data: 0.9069  max mem: 17867
[01:08:33.777185] Epoch: [70]  [2000/5004]  eta: 0:17:31  lr: 0.000897  loss: 2.9766 (3.1262)  time: 0.3473  data: 0.0002  max mem: 17867
[01:20:12.363770] Epoch: [70]  [4000/5004]  eta: 0:05:51  lr: 0.000894  loss: 3.0404 (3.1380)  time: 0.3485  data: 0.0002  max mem: 17867
[01:26:02.728832] Epoch: [70]  [5003/5004]  eta: 0:00:00  lr: 0.000892  loss: 2.9178 (3.1418)  time: 0.3442  data: 0.0011  max mem: 17867
[01:26:03.135663] Epoch: [70] Total time: 0:29:09 (0.3497 s / it)
[01:26:03.136378] Averaged stats: lr: 0.000892  loss: 2.9178 (3.1447)
[01:26:04.406762] Test:  [   0/1563]  eta: 0:33:00  loss: 0.6205 (0.6205)  acc1: 84.3750 (84.3750)  acc5: 96.8750 (96.8750)  time: 1.2669  data: 1.1236  max mem: 17867
[01:27:09.387622] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.9639 (1.0772)  acc1: 78.1250 (72.8668)  acc5: 96.8750 (93.3321)  time: 0.1298  data: 0.0002  max mem: 17867
[01:28:14.389406] Test:  [1000/1563]  eta: 0:01:13  loss: 1.2316 (1.2217)  acc1: 68.7500 (70.0206)  acc5: 90.6250 (90.9403)  time: 0.1299  data: 0.0002  max mem: 17867
[01:29:19.330680] Test:  [1500/1563]  eta: 0:00:08  loss: 0.9784 (1.3063)  acc1: 78.1250 (68.3773)  acc5: 90.6250 (89.5237)  time: 0.1298  data: 0.0002  max mem: 17867
[01:29:27.308400] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4994 (1.3063)  acc1: 87.5000 (68.3740)  acc5: 96.8750 (89.5340)  time: 0.1261  data: 0.0001  max mem: 17867
[01:29:27.383326] Test: Total time: 0:03:24 (0.1307 s / it)
[01:29:27.719510] * Acc@1 68.374 Acc@5 89.534 loss 1.306
[01:29:27.719673] Accuracy of the network on the 50000 test images: 68.4%
[01:29:27.719695] Max accuracy: 68.37%
[01:29:27.728116] log_dir: ./output_dir_cml_spikformer
[01:29:29.230670] Epoch: [71]  [   0/5004]  eta: 2:05:15  lr: 0.000892  loss: 3.1025 (3.1025)  time: 1.5019  data: 0.9741  max mem: 17867
[01:41:08.361497] Epoch: [71]  [2000/5004]  eta: 0:17:31  lr: 0.000888  loss: 3.0260 (3.1159)  time: 0.3479  data: 0.0002  max mem: 17867
[01:52:47.079329] Epoch: [71]  [4000/5004]  eta: 0:05:51  lr: 0.000885  loss: 3.0837 (3.1247)  time: 0.3484  data: 0.0002  max mem: 17867
[01:58:37.403325] Epoch: [71]  [5003/5004]  eta: 0:00:00  lr: 0.000883  loss: 3.0941 (3.1297)  time: 0.3496  data: 0.0006  max mem: 17867
[01:58:37.753507] Epoch: [71] Total time: 0:29:10 (0.3497 s / it)
[01:58:37.777997] Averaged stats: lr: 0.000883  loss: 3.0941 (3.1373)
[01:58:39.007191] Test:  [   0/1563]  eta: 0:31:54  loss: 0.5247 (0.5247)  acc1: 90.6250 (90.6250)  acc5: 96.8750 (96.8750)  time: 1.2246  data: 1.0194  max mem: 17867
[01:59:44.033810] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.0071 (1.0915)  acc1: 75.0000 (72.8169)  acc5: 93.7500 (93.1637)  time: 0.1299  data: 0.0002  max mem: 17867
[02:00:49.092374] Test:  [1000/1563]  eta: 0:01:13  loss: 1.4985 (1.2424)  acc1: 56.2500 (69.8114)  acc5: 90.6250 (90.6687)  time: 0.1301  data: 0.0002  max mem: 17867
[02:01:54.099519] Test:  [1500/1563]  eta: 0:00:08  loss: 0.7396 (1.3314)  acc1: 81.2500 (67.9984)  acc5: 93.7500 (89.3717)  time: 0.1300  data: 0.0002  max mem: 17867
[02:02:02.087649] Test:  [1562/1563]  eta: 0:00:00  loss: 0.6592 (1.3296)  acc1: 87.5000 (68.0480)  acc5: 96.8750 (89.3900)  time: 0.1262  data: 0.0001  max mem: 17867
[02:02:02.161864] Test: Total time: 0:03:24 (0.1308 s / it)
[02:02:02.162893] * Acc@1 68.048 Acc@5 89.390 loss 1.330
[02:02:02.163003] Accuracy of the network on the 50000 test images: 68.0%
[02:02:02.163024] Max accuracy: 68.37%
[02:02:02.171683] log_dir: ./output_dir_cml_spikformer
[02:02:03.614130] Epoch: [72]  [   0/5004]  eta: 2:00:14  lr: 0.000883  loss: 3.5137 (3.5137)  time: 1.4417  data: 0.9965  max mem: 17867
[02:13:43.168984] Epoch: [72]  [2000/5004]  eta: 0:17:32  lr: 0.000880  loss: 2.9596 (3.1110)  time: 0.3488  data: 0.0002  max mem: 17867
[02:25:22.794743] Epoch: [72]  [4000/5004]  eta: 0:05:51  lr: 0.000877  loss: 3.0175 (3.1265)  time: 0.3495  data: 0.0003  max mem: 17867
[02:31:13.142090] Epoch: [72]  [5003/5004]  eta: 0:00:00  lr: 0.000875  loss: 3.1637 (3.1303)  time: 0.3494  data: 0.0012  max mem: 17867
[02:31:13.502549] Epoch: [72] Total time: 0:29:11 (0.3500 s / it)
[02:31:13.503455] Averaged stats: lr: 0.000875  loss: 3.1637 (3.1361)
[02:31:14.463726] Test:  [   0/1563]  eta: 0:24:53  loss: 0.3819 (0.3819)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 0.9554  data: 0.8064  max mem: 17867
[02:32:19.590656] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.2553 (1.0809)  acc1: 65.6250 (73.4032)  acc5: 90.6250 (93.2136)  time: 0.1299  data: 0.0002  max mem: 17867
[02:33:24.579158] Test:  [1000/1563]  eta: 0:01:13  loss: 1.4579 (1.2253)  acc1: 62.5000 (70.3609)  acc5: 87.5000 (90.9871)  time: 0.1299  data: 0.0002  max mem: 17867
[02:34:29.576684] Test:  [1500/1563]  eta: 0:00:08  loss: 0.7754 (1.3208)  acc1: 84.3750 (68.3149)  acc5: 90.6250 (89.5091)  time: 0.1300  data: 0.0003  max mem: 17867
[02:34:37.567446] Test:  [1562/1563]  eta: 0:00:00  loss: 0.5253 (1.3207)  acc1: 90.6250 (68.3560)  acc5: 96.8750 (89.5140)  time: 0.1262  data: 0.0001  max mem: 17867
[02:34:37.633345] Test: Total time: 0:03:24 (0.1306 s / it)
[02:34:37.792981] * Acc@1 68.356 Acc@5 89.514 loss 1.321
[02:34:37.793118] Accuracy of the network on the 50000 test images: 68.4%
[02:34:37.793138] Max accuracy: 68.37%
[02:34:37.815185] log_dir: ./output_dir_cml_spikformer
[02:34:39.380843] Epoch: [73]  [   0/5004]  eta: 2:10:29  lr: 0.000875  loss: 3.0208 (3.0208)  time: 1.5647  data: 1.1194  max mem: 17867
[02:46:21.209169] Epoch: [73]  [2000/5004]  eta: 0:17:35  lr: 0.000871  loss: 3.0538 (3.1173)  time: 0.3545  data: 0.0002  max mem: 17867
[02:58:01.836942] Epoch: [73]  [4000/5004]  eta: 0:05:52  lr: 0.000868  loss: 3.2919 (3.1307)  time: 0.3474  data: 0.0002  max mem: 17867
[03:03:53.345210] Epoch: [73]  [5003/5004]  eta: 0:00:00  lr: 0.000866  loss: 3.0404 (3.1297)  time: 0.3479  data: 0.0011  max mem: 17867
[03:03:53.721271] Epoch: [73] Total time: 0:29:15 (0.3509 s / it)
[03:03:53.724194] Averaged stats: lr: 0.000866  loss: 3.0404 (3.1273)
[03:03:54.762853] Test:  [   0/1563]  eta: 0:26:29  loss: 0.5503 (0.5503)  acc1: 87.5000 (87.5000)  acc5: 96.8750 (96.8750)  time: 1.0170  data: 0.8787  max mem: 17867
[03:04:59.752932] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.0744 (0.9854)  acc1: 68.7500 (74.3825)  acc5: 93.7500 (94.1679)  time: 0.1299  data: 0.0002  max mem: 17867
[03:06:04.691444] Test:  [1000/1563]  eta: 0:01:13  loss: 1.5769 (1.1457)  acc1: 59.3750 (71.4130)  acc5: 90.6250 (91.5460)  time: 0.1299  data: 0.0002  max mem: 17867
[03:07:09.681441] Test:  [1500/1563]  eta: 0:00:08  loss: 0.8774 (1.2467)  acc1: 81.2500 (69.3912)  acc5: 93.7500 (90.1940)  time: 0.1298  data: 0.0002  max mem: 17867
[03:07:17.664553] Test:  [1562/1563]  eta: 0:00:00  loss: 0.5377 (1.2459)  acc1: 87.5000 (69.3600)  acc5: 96.8750 (90.1940)  time: 0.1261  data: 0.0001  max mem: 17867
[03:07:17.723728] Test: Total time: 0:03:23 (0.1305 s / it)
[03:07:17.978959] * Acc@1 69.360 Acc@5 90.194 loss 1.246
[03:07:17.979134] Accuracy of the network on the 50000 test images: 69.4%
[03:07:17.979161] Max accuracy: 69.36%
[03:07:18.008194] log_dir: ./output_dir_cml_spikformer
[03:07:19.466478] Epoch: [74]  [   0/5004]  eta: 2:01:33  lr: 0.000866  loss: 2.7412 (2.7412)  time: 1.4575  data: 0.9413  max mem: 17867
[03:18:59.862403] Epoch: [74]  [2000/5004]  eta: 0:17:33  lr: 0.000863  loss: 3.0791 (3.1303)  time: 0.3515  data: 0.0002  max mem: 17867
[03:30:40.005137] Epoch: [74]  [4000/5004]  eta: 0:05:51  lr: 0.000859  loss: 3.1411 (3.1282)  time: 0.3477  data: 0.0002  max mem: 17867
[03:36:31.301024] Epoch: [74]  [5003/5004]  eta: 0:00:00  lr: 0.000858  loss: 3.1785 (3.1309)  time: 0.3456  data: 0.0006  max mem: 17867
[03:36:31.659578] Epoch: [74] Total time: 0:29:13 (0.3504 s / it)
[03:36:31.669415] Averaged stats: lr: 0.000858  loss: 3.1785 (3.1230)
[03:36:32.717508] Test:  [   0/1563]  eta: 0:27:11  loss: 0.5003 (0.5003)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.0439  data: 0.9059  max mem: 17867
[03:37:37.885365] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.3373 (1.0442)  acc1: 68.7500 (73.4406)  acc5: 93.7500 (93.9309)  time: 0.1300  data: 0.0002  max mem: 17867
[03:38:42.857790] Test:  [1000/1563]  eta: 0:01:13  loss: 1.6090 (1.2066)  acc1: 53.1250 (70.4077)  acc5: 87.5000 (91.2275)  time: 0.1299  data: 0.0002  max mem: 17867
[03:39:47.853789] Test:  [1500/1563]  eta: 0:00:08  loss: 0.8522 (1.3089)  acc1: 81.2500 (68.4335)  acc5: 93.7500 (89.6173)  time: 0.1299  data: 0.0002  max mem: 17867
[03:39:55.874108] Test:  [1562/1563]  eta: 0:00:00  loss: 0.5462 (1.3066)  acc1: 87.5000 (68.4960)  acc5: 93.7500 (89.6140)  time: 0.1262  data: 0.0001  max mem: 17867
[03:39:55.956730] Test: Total time: 0:03:24 (0.1307 s / it)
[03:39:56.313877] * Acc@1 68.496 Acc@5 89.614 loss 1.307
[03:39:56.314019] Accuracy of the network on the 50000 test images: 68.5%
[03:39:56.314041] Max accuracy: 69.36%
[03:39:56.339727] log_dir: ./output_dir_cml_spikformer
[03:39:57.880837] Epoch: [75]  [   0/5004]  eta: 2:08:26  lr: 0.000858  loss: 3.3385 (3.3385)  time: 1.5401  data: 0.9880  max mem: 17867
[03:51:37.872884] Epoch: [75]  [2000/5004]  eta: 0:17:33  lr: 0.000854  loss: 3.0585 (3.1158)  time: 0.3517  data: 0.0002  max mem: 17867
[04:03:17.738660] Epoch: [75]  [4000/5004]  eta: 0:05:51  lr: 0.000851  loss: 3.0782 (3.1134)  time: 0.3472  data: 0.0002  max mem: 17867
[04:09:09.042095] Epoch: [75]  [5003/5004]  eta: 0:00:00  lr: 0.000849  loss: 2.9357 (3.1183)  time: 0.3458  data: 0.0006  max mem: 17867
[04:09:09.424849] Epoch: [75] Total time: 0:29:13 (0.3503 s / it)
[04:09:09.430297] Averaged stats: lr: 0.000849  loss: 2.9357 (3.1204)
[04:09:10.407523] Test:  [   0/1563]  eta: 0:25:21  loss: 0.3571 (0.3571)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 0.9737  data: 0.8349  max mem: 17867
[04:10:15.546443] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.0392 (1.0506)  acc1: 68.7500 (73.3221)  acc5: 93.7500 (93.3383)  time: 0.1299  data: 0.0002  max mem: 17867
[04:11:20.539169] Test:  [1000/1563]  eta: 0:01:13  loss: 1.3847 (1.1945)  acc1: 68.7500 (70.3640)  acc5: 87.5000 (91.1120)  time: 0.1299  data: 0.0002  max mem: 17867
[04:12:25.526312] Test:  [1500/1563]  eta: 0:00:08  loss: 0.7606 (1.2874)  acc1: 81.2500 (68.5002)  acc5: 93.7500 (89.5757)  time: 0.1299  data: 0.0002  max mem: 17867
[04:12:33.511063] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4121 (1.2846)  acc1: 90.6250 (68.5620)  acc5: 96.8750 (89.6380)  time: 0.1262  data: 0.0001  max mem: 17867
[04:12:33.574605] Test: Total time: 0:03:24 (0.1306 s / it)
[04:12:34.045965] * Acc@1 68.562 Acc@5 89.638 loss 1.285
[04:12:34.046113] Accuracy of the network on the 50000 test images: 68.6%
[04:12:34.046135] Max accuracy: 69.36%
[04:12:34.053101] log_dir: ./output_dir_cml_spikformer
[04:12:35.553390] Epoch: [76]  [   0/5004]  eta: 2:05:02  lr: 0.000849  loss: 2.9123 (2.9123)  time: 1.4992  data: 0.9810  max mem: 17867
[04:24:15.551429] Epoch: [76]  [2000/5004]  eta: 0:17:33  lr: 0.000845  loss: 3.1643 (3.1059)  time: 0.3489  data: 0.0002  max mem: 17867
[04:35:53.924182] Epoch: [76]  [4000/5004]  eta: 0:05:51  lr: 0.000842  loss: 3.0431 (3.1146)  time: 0.3459  data: 0.0002  max mem: 17867
[04:41:44.659824] Epoch: [76]  [5003/5004]  eta: 0:00:00  lr: 0.000840  loss: 3.1193 (3.1139)  time: 0.3468  data: 0.0011  max mem: 17867
[04:41:45.009951] Epoch: [76] Total time: 0:29:10 (0.3499 s / it)
[04:41:45.019836] Averaged stats: lr: 0.000840  loss: 3.1193 (3.1112)
[04:41:46.054843] Test:  [   0/1563]  eta: 0:26:52  loss: 0.4883 (0.4883)  acc1: 90.6250 (90.6250)  acc5: 96.8750 (96.8750)  time: 1.0316  data: 0.8653  max mem: 17867
[04:42:51.045456] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.1289 (1.0326)  acc1: 65.6250 (73.2597)  acc5: 93.7500 (93.9122)  time: 0.1299  data: 0.0002  max mem: 17867
[04:43:56.045720] Test:  [1000/1563]  eta: 0:01:13  loss: 1.8308 (1.2045)  acc1: 53.1250 (70.2485)  acc5: 81.2500 (91.1495)  time: 0.1299  data: 0.0002  max mem: 17867
[04:45:01.027414] Test:  [1500/1563]  eta: 0:00:08  loss: 0.6555 (1.3003)  acc1: 81.2500 (68.3232)  acc5: 96.8750 (89.6944)  time: 0.1299  data: 0.0002  max mem: 17867
[04:45:09.010193] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4186 (1.2993)  acc1: 90.6250 (68.3200)  acc5: 96.8750 (89.7140)  time: 0.1262  data: 0.0001  max mem: 17867
[04:45:09.083006] Test: Total time: 0:03:24 (0.1306 s / it)
[04:45:09.225888] * Acc@1 68.320 Acc@5 89.714 loss 1.299
[04:45:09.226029] Accuracy of the network on the 50000 test images: 68.3%
[04:45:09.226049] Max accuracy: 69.36%
[04:45:09.282850] log_dir: ./output_dir_cml_spikformer
[04:45:10.726704] Epoch: [77]  [   0/5004]  eta: 2:00:20  lr: 0.000840  loss: 2.6077 (2.6077)  time: 1.4430  data: 1.0650  max mem: 17867
[04:56:50.366711] Epoch: [77]  [2000/5004]  eta: 0:17:32  lr: 0.000836  loss: 3.2352 (3.0939)  time: 0.3465  data: 0.0002  max mem: 17867
[05:08:29.033598] Epoch: [77]  [4000/5004]  eta: 0:05:51  lr: 0.000833  loss: 3.0015 (3.1012)  time: 0.3458  data: 0.0002  max mem: 17867
[05:14:19.289165] Epoch: [77]  [5003/5004]  eta: 0:00:00  lr: 0.000831  loss: 3.3241 (3.1099)  time: 0.3446  data: 0.0011  max mem: 17867
[05:14:19.645245] Epoch: [77] Total time: 0:29:10 (0.3498 s / it)
[05:14:19.646193] Averaged stats: lr: 0.000831  loss: 3.3241 (3.1129)
[05:14:20.717109] Test:  [   0/1563]  eta: 0:27:47  loss: 0.3557 (0.3557)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.0670  data: 0.9172  max mem: 17867
[05:15:25.735603] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.0083 (1.0251)  acc1: 71.8750 (73.3907)  acc5: 93.7500 (93.8373)  time: 0.1300  data: 0.0002  max mem: 17867
[05:16:30.729394] Test:  [1000/1563]  eta: 0:01:13  loss: 1.4807 (1.1758)  acc1: 56.2500 (70.9416)  acc5: 90.6250 (91.3493)  time: 0.1300  data: 0.0002  max mem: 17867
[05:17:35.754199] Test:  [1500/1563]  eta: 0:00:08  loss: 0.7937 (1.2734)  acc1: 81.2500 (68.7250)  acc5: 93.7500 (89.8526)  time: 0.1301  data: 0.0002  max mem: 17867
[05:17:43.745075] Test:  [1562/1563]  eta: 0:00:00  loss: 0.5395 (1.2753)  acc1: 87.5000 (68.6840)  acc5: 96.8750 (89.8460)  time: 0.1263  data: 0.0001  max mem: 17867
[05:17:43.830609] Test: Total time: 0:03:24 (0.1306 s / it)
[05:17:44.058137] * Acc@1 68.684 Acc@5 89.846 loss 1.275
[05:17:44.058298] Accuracy of the network on the 50000 test images: 68.7%
[05:17:44.058321] Max accuracy: 69.36%
[05:17:44.084196] log_dir: ./output_dir_cml_spikformer
[05:17:45.513817] Epoch: [78]  [   0/5004]  eta: 1:59:09  lr: 0.000831  loss: 3.0864 (3.0864)  time: 1.4288  data: 0.9088  max mem: 17867
[05:29:25.524339] Epoch: [78]  [2000/5004]  eta: 0:17:32  lr: 0.000827  loss: 2.9177 (3.1043)  time: 0.3512  data: 0.0002  max mem: 17867
[05:41:04.406380] Epoch: [78]  [4000/5004]  eta: 0:05:51  lr: 0.000824  loss: 3.1287 (3.1031)  time: 0.3490  data: 0.0002  max mem: 17867
[05:46:54.924517] Epoch: [78]  [5003/5004]  eta: 0:00:00  lr: 0.000822  loss: 3.0343 (3.1067)  time: 0.3488  data: 0.0012  max mem: 17867
[05:46:55.300297] Epoch: [78] Total time: 0:29:11 (0.3500 s / it)
[05:46:55.304291] Averaged stats: lr: 0.000822  loss: 3.0343 (3.1046)
[05:46:56.607045] Test:  [   0/1563]  eta: 0:33:50  loss: 0.3093 (0.3093)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.2992  data: 0.9381  max mem: 17867
[05:48:01.576660] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.0052 (1.0514)  acc1: 75.0000 (73.2223)  acc5: 93.7500 (93.7375)  time: 0.1298  data: 0.0002  max mem: 17867
[05:49:06.703251] Test:  [1000/1563]  eta: 0:01:13  loss: 1.4050 (1.2187)  acc1: 59.3750 (69.8895)  acc5: 90.6250 (91.1901)  time: 0.1302  data: 0.0002  max mem: 17867
[05:50:11.736287] Test:  [1500/1563]  eta: 0:00:08  loss: 0.6622 (1.2999)  acc1: 84.3750 (68.5002)  acc5: 96.8750 (89.9275)  time: 0.1298  data: 0.0002  max mem: 17867
[05:50:19.715551] Test:  [1562/1563]  eta: 0:00:00  loss: 0.6380 (1.3002)  acc1: 87.5000 (68.5100)  acc5: 96.8750 (89.9120)  time: 0.1262  data: 0.0001  max mem: 17867
[05:50:19.794876] Test: Total time: 0:03:24 (0.1308 s / it)
[05:50:20.127009] * Acc@1 68.510 Acc@5 89.912 loss 1.300
[05:50:20.127174] Accuracy of the network on the 50000 test images: 68.5%
[05:50:20.127197] Max accuracy: 69.36%
[05:50:20.147633] log_dir: ./output_dir_cml_spikformer
[05:50:21.587450] Epoch: [79]  [   0/5004]  eta: 2:00:00  lr: 0.000822  loss: 2.8981 (2.8981)  time: 1.4390  data: 1.0939  max mem: 17867
[06:02:01.202183] Epoch: [79]  [2000/5004]  eta: 0:17:32  lr: 0.000818  loss: 3.0721 (3.0845)  time: 0.3470  data: 0.0002  max mem: 17867
[06:13:40.919422] Epoch: [79]  [4000/5004]  eta: 0:05:51  lr: 0.000815  loss: 2.8963 (3.0922)  time: 0.3464  data: 0.0002  max mem: 17867
[06:19:32.337566] Epoch: [79]  [5003/5004]  eta: 0:00:00  lr: 0.000813  loss: 3.0960 (3.0952)  time: 0.3443  data: 0.0011  max mem: 17867
[06:19:32.675381] Epoch: [79] Total time: 0:29:12 (0.3502 s / it)
[06:19:32.685311] Averaged stats: lr: 0.000813  loss: 3.0960 (3.1000)
[06:19:33.649491] Test:  [   0/1563]  eta: 0:25:01  loss: 0.4916 (0.4916)  acc1: 93.7500 (93.7500)  acc5: 93.7500 (93.7500)  time: 0.9606  data: 0.8230  max mem: 17867
[06:20:38.653058] Test:  [ 500/1563]  eta: 0:02:19  loss: 0.9260 (0.9895)  acc1: 75.0000 (73.8398)  acc5: 93.7500 (94.1866)  time: 0.1299  data: 0.0002  max mem: 17867
[06:21:43.649973] Test:  [1000/1563]  eta: 0:01:13  loss: 1.5153 (1.1494)  acc1: 59.3750 (71.0758)  acc5: 87.5000 (91.7395)  time: 0.1299  data: 0.0002  max mem: 17867
[06:22:48.638159] Test:  [1500/1563]  eta: 0:00:08  loss: 0.6665 (1.2471)  acc1: 84.3750 (69.2268)  acc5: 96.8750 (90.2502)  time: 0.1299  data: 0.0002  max mem: 17867
[06:22:56.624772] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4776 (1.2473)  acc1: 84.3750 (69.2840)  acc5: 96.8750 (90.2520)  time: 0.1263  data: 0.0001  max mem: 17867
[06:22:56.700820] Test: Total time: 0:03:24 (0.1305 s / it)
[06:22:57.001327] * Acc@1 69.284 Acc@5 90.252 loss 1.247
[06:22:57.001468] Accuracy of the network on the 50000 test images: 69.3%
[06:22:57.001490] Max accuracy: 69.36%
[06:22:57.008523] log_dir: ./output_dir_cml_spikformer
[06:22:58.429815] Epoch: [80]  [   0/5004]  eta: 1:58:28  lr: 0.000813  loss: 2.7097 (2.7097)  time: 1.4205  data: 1.0013  max mem: 17867
[06:34:39.741990] Epoch: [80]  [2000/5004]  eta: 0:17:34  lr: 0.000809  loss: 2.9372 (3.0685)  time: 0.3536  data: 0.0002  max mem: 17867
[06:46:19.600138] Epoch: [80]  [4000/5004]  eta: 0:05:51  lr: 0.000806  loss: 3.0704 (3.0741)  time: 0.3441  data: 0.0002  max mem: 17867
[06:52:10.317158] Epoch: [80]  [5003/5004]  eta: 0:00:00  lr: 0.000804  loss: 3.0652 (3.0757)  time: 0.3458  data: 0.0011  max mem: 17867
[06:52:10.684419] Epoch: [80] Total time: 0:29:13 (0.3505 s / it)
[06:52:10.685410] Averaged stats: lr: 0.000804  loss: 3.0652 (3.0922)
[06:52:12.004631] Test:  [   0/1563]  eta: 0:34:16  loss: 0.4283 (0.4283)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.3157  data: 1.1752  max mem: 17867
[06:53:17.022112] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.9044 (1.0579)  acc1: 71.8750 (73.6215)  acc5: 93.7500 (93.6315)  time: 0.1298  data: 0.0002  max mem: 17867
[06:54:21.960256] Test:  [1000/1563]  eta: 0:01:13  loss: 1.5434 (1.2077)  acc1: 59.3750 (70.5825)  acc5: 87.5000 (91.2369)  time: 0.1298  data: 0.0002  max mem: 17867
[06:55:26.925444] Test:  [1500/1563]  eta: 0:00:08  loss: 0.7566 (1.2890)  acc1: 84.3750 (68.8812)  acc5: 96.8750 (89.9046)  time: 0.1299  data: 0.0002  max mem: 17867
[06:55:34.903755] Test:  [1562/1563]  eta: 0:00:00  loss: 0.5476 (1.2888)  acc1: 90.6250 (68.8760)  acc5: 96.8750 (89.9320)  time: 0.1261  data: 0.0001  max mem: 17867
[06:55:35.003272] Test: Total time: 0:03:24 (0.1307 s / it)
[06:55:35.147641] * Acc@1 68.876 Acc@5 89.932 loss 1.289
[06:55:35.147801] Accuracy of the network on the 50000 test images: 68.9%
[06:55:35.147824] Max accuracy: 69.36%
[06:55:35.155161] log_dir: ./output_dir_cml_spikformer
[06:55:36.735392] Epoch: [81]  [   0/5004]  eta: 2:11:44  lr: 0.000804  loss: 2.8942 (2.8942)  time: 1.5796  data: 1.0055  max mem: 17867
[07:07:17.294020] Epoch: [81]  [2000/5004]  eta: 0:17:34  lr: 0.000800  loss: 3.1493 (3.0926)  time: 0.3538  data: 0.0002  max mem: 17867
[07:18:57.222365] Epoch: [81]  [4000/5004]  eta: 0:05:51  lr: 0.000797  loss: 3.0710 (3.0927)  time: 0.3526  data: 0.0002  max mem: 17867
[07:24:48.159505] Epoch: [81]  [5003/5004]  eta: 0:00:00  lr: 0.000795  loss: 3.1054 (3.0933)  time: 0.3455  data: 0.0006  max mem: 17867
[07:24:48.551931] Epoch: [81] Total time: 0:29:13 (0.3504 s / it)
[07:24:48.558101] Averaged stats: lr: 0.000795  loss: 3.1054 (3.0885)
[07:24:49.587497] Test:  [   0/1563]  eta: 0:26:43  loss: 0.3527 (0.3527)  acc1: 90.6250 (90.6250)  acc5: 96.8750 (96.8750)  time: 1.0259  data: 0.8508  max mem: 17867
[07:25:54.639918] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.8607 (0.9649)  acc1: 75.0000 (74.7630)  acc5: 96.8750 (94.1804)  time: 0.1309  data: 0.0002  max mem: 17867
[07:26:59.671307] Test:  [1000/1563]  eta: 0:01:13  loss: 1.5675 (1.1193)  acc1: 62.5000 (71.6502)  acc5: 87.5000 (91.8644)  time: 0.1299  data: 0.0002  max mem: 17867
[07:28:04.650593] Test:  [1500/1563]  eta: 0:00:08  loss: 0.6455 (1.2176)  acc1: 84.3750 (69.6223)  acc5: 96.8750 (90.5542)  time: 0.1299  data: 0.0002  max mem: 17867
[07:28:12.633814] Test:  [1562/1563]  eta: 0:00:00  loss: 0.5465 (1.2199)  acc1: 87.5000 (69.5780)  acc5: 96.8750 (90.4980)  time: 0.1263  data: 0.0001  max mem: 17867
[07:28:12.717698] Test: Total time: 0:03:24 (0.1306 s / it)
[07:28:13.008068] * Acc@1 69.578 Acc@5 90.498 loss 1.220
[07:28:13.008201] Accuracy of the network on the 50000 test images: 69.6%
[07:28:13.008219] Max accuracy: 69.58%
[07:28:13.034825] log_dir: ./output_dir_cml_spikformer
[07:28:14.471366] Epoch: [82]  [   0/5004]  eta: 1:59:44  lr: 0.000795  loss: 3.1227 (3.1227)  time: 1.4358  data: 0.9828  max mem: 17867
[07:39:54.045599] Epoch: [82]  [2000/5004]  eta: 0:17:32  lr: 0.000791  loss: 2.9133 (3.0625)  time: 0.3534  data: 0.0002  max mem: 17867
[07:51:33.392246] Epoch: [82]  [4000/5004]  eta: 0:05:51  lr: 0.000788  loss: 3.2185 (3.0692)  time: 0.3488  data: 0.0002  max mem: 17867
[07:57:23.467974] Epoch: [82]  [5003/5004]  eta: 0:00:00  lr: 0.000786  loss: 3.1264 (3.0786)  time: 0.3453  data: 0.0011  max mem: 17867
[07:57:23.813648] Epoch: [82] Total time: 0:29:10 (0.3499 s / it)
[07:57:23.823341] Averaged stats: lr: 0.000786  loss: 3.1264 (3.0841)
[07:57:24.872341] Test:  [   0/1563]  eta: 0:27:14  loss: 0.4867 (0.4867)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.0454  data: 0.9103  max mem: 17867
[07:58:30.062445] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.9218 (1.0341)  acc1: 71.8750 (74.0145)  acc5: 93.7500 (93.6315)  time: 0.1300  data: 0.0002  max mem: 17867
[07:59:35.061489] Test:  [1000/1563]  eta: 0:01:13  loss: 1.5143 (1.1809)  acc1: 56.2500 (70.9509)  acc5: 90.6250 (91.5054)  time: 0.1299  data: 0.0002  max mem: 17867
[08:00:40.065822] Test:  [1500/1563]  eta: 0:00:08  loss: 0.7369 (1.2776)  acc1: 84.3750 (68.9894)  acc5: 96.8750 (90.0337)  time: 0.1299  data: 0.0002  max mem: 17867
[08:00:48.046657] Test:  [1562/1563]  eta: 0:00:00  loss: 0.6329 (1.2774)  acc1: 87.5000 (68.9740)  acc5: 96.8750 (90.0460)  time: 0.1262  data: 0.0001  max mem: 17867
[08:00:48.111597] Test: Total time: 0:03:24 (0.1307 s / it)
[08:00:48.632795] * Acc@1 68.974 Acc@5 90.046 loss 1.277
[08:00:48.632931] Accuracy of the network on the 50000 test images: 69.0%
[08:00:48.632953] Max accuracy: 69.58%
[08:00:48.651327] log_dir: ./output_dir_cml_spikformer
[08:00:50.095249] Epoch: [83]  [   0/5004]  eta: 2:00:08  lr: 0.000786  loss: 2.9808 (2.9808)  time: 1.4405  data: 1.0595  max mem: 17867
[08:12:29.851606] Epoch: [83]  [2000/5004]  eta: 0:17:32  lr: 0.000782  loss: 2.9160 (3.0724)  time: 0.3500  data: 0.0002  max mem: 17867
[08:24:09.211177] Epoch: [83]  [4000/5004]  eta: 0:05:51  lr: 0.000778  loss: 3.0038 (3.0766)  time: 0.3473  data: 0.0002  max mem: 17867
[08:29:59.710288] Epoch: [83]  [5003/5004]  eta: 0:00:00  lr: 0.000777  loss: 3.0072 (3.0808)  time: 0.3444  data: 0.0011  max mem: 17867
[08:30:00.079203] Epoch: [83] Total time: 0:29:11 (0.3500 s / it)
[08:30:00.085144] Averaged stats: lr: 0.000777  loss: 3.0072 (3.0778)
[08:30:01.131832] Test:  [   0/1563]  eta: 0:27:08  loss: 0.4767 (0.4767)  acc1: 90.6250 (90.6250)  acc5: 96.8750 (96.8750)  time: 1.0416  data: 0.8882  max mem: 17867
[08:31:06.160250] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.9694 (1.0154)  acc1: 75.0000 (74.6382)  acc5: 93.7500 (93.9122)  time: 0.1300  data: 0.0002  max mem: 17867
[08:32:11.147614] Test:  [1000/1563]  eta: 0:01:13  loss: 1.2105 (1.1643)  acc1: 68.7500 (71.4535)  acc5: 90.6250 (91.6334)  time: 0.1299  data: 0.0002  max mem: 17867
[08:33:16.189035] Test:  [1500/1563]  eta: 0:00:08  loss: 0.7228 (1.2663)  acc1: 84.3750 (69.1872)  acc5: 96.8750 (90.0150)  time: 0.1299  data: 0.0002  max mem: 17867
[08:33:24.173236] Test:  [1562/1563]  eta: 0:00:00  loss: 0.5592 (1.2644)  acc1: 87.5000 (69.2560)  acc5: 96.8750 (90.0520)  time: 0.1262  data: 0.0001  max mem: 17867
[08:33:24.257257] Test: Total time: 0:03:24 (0.1306 s / it)
[08:33:24.511845] * Acc@1 69.256 Acc@5 90.052 loss 1.264
[08:33:24.512019] Accuracy of the network on the 50000 test images: 69.3%
[08:33:24.512044] Max accuracy: 69.58%
[08:33:24.518659] log_dir: ./output_dir_cml_spikformer
[08:33:25.981337] Epoch: [84]  [   0/5004]  eta: 2:01:53  lr: 0.000777  loss: 3.6815 (3.6815)  time: 1.4616  data: 1.1217  max mem: 17867
[08:45:06.047283] Epoch: [84]  [2000/5004]  eta: 0:17:33  lr: 0.000773  loss: 2.9994 (3.0644)  time: 0.3446  data: 0.0002  max mem: 17867
[08:56:45.651678] Epoch: [84]  [4000/5004]  eta: 0:05:51  lr: 0.000769  loss: 3.0884 (3.0737)  time: 0.3497  data: 0.0002  max mem: 17867
[09:02:36.291371] Epoch: [84]  [5003/5004]  eta: 0:00:00  lr: 0.000767  loss: 2.8436 (3.0763)  time: 0.3435  data: 0.0006  max mem: 17867
[09:02:36.642875] Epoch: [84] Total time: 0:29:12 (0.3501 s / it)
[09:02:36.649354] Averaged stats: lr: 0.000767  loss: 2.8436 (3.0745)
[09:02:37.730739] Test:  [   0/1563]  eta: 0:28:02  loss: 0.3992 (0.3992)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.0766  data: 0.8536  max mem: 17867
[09:03:42.713387] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.1517 (1.0062)  acc1: 65.6250 (73.8273)  acc5: 93.7500 (94.0120)  time: 0.1299  data: 0.0002  max mem: 17867
[09:04:47.738557] Test:  [1000/1563]  eta: 0:01:13  loss: 1.3083 (1.1605)  acc1: 65.6250 (71.0227)  acc5: 90.6250 (91.5522)  time: 0.1300  data: 0.0002  max mem: 17867
[09:05:53.230999] Test:  [1500/1563]  eta: 0:00:08  loss: 0.7385 (1.2578)  acc1: 84.3750 (69.0019)  acc5: 96.8750 (90.1191)  time: 0.1299  data: 0.0002  max mem: 17867
[09:06:01.217598] Test:  [1562/1563]  eta: 0:00:00  loss: 0.6233 (1.2528)  acc1: 87.5000 (69.0940)  acc5: 96.8750 (90.1920)  time: 0.1262  data: 0.0001  max mem: 17867
[09:06:01.288361] Test: Total time: 0:03:24 (0.1309 s / it)
[09:06:01.459737] * Acc@1 69.094 Acc@5 90.192 loss 1.253
[09:06:01.459888] Accuracy of the network on the 50000 test images: 69.1%
[09:06:01.459909] Max accuracy: 69.58%
[09:06:01.479079] log_dir: ./output_dir_cml_spikformer
[09:06:02.966542] Epoch: [85]  [   0/5004]  eta: 2:03:59  lr: 0.000767  loss: 3.0865 (3.0865)  time: 1.4868  data: 1.1016  max mem: 17867
[09:17:43.157951] Epoch: [85]  [2000/5004]  eta: 0:17:33  lr: 0.000764  loss: 3.0073 (3.0543)  time: 0.3478  data: 0.0002  max mem: 17867
[09:29:22.902270] Epoch: [85]  [4000/5004]  eta: 0:05:51  lr: 0.000760  loss: 2.9964 (3.0607)  time: 0.3479  data: 0.0002  max mem: 17867
[09:35:14.144983] Epoch: [85]  [5003/5004]  eta: 0:00:00  lr: 0.000758  loss: 2.9272 (3.0618)  time: 0.3495  data: 0.0012  max mem: 17867
[09:35:14.498645] Epoch: [85] Total time: 0:29:13 (0.3503 s / it)
[09:35:14.507430] Averaged stats: lr: 0.000758  loss: 2.9272 (3.0671)
[09:35:15.476153] Test:  [   0/1563]  eta: 0:25:08  loss: 0.3546 (0.3546)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 0.9650  data: 0.8166  max mem: 17867
[09:36:20.640235] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.2068 (1.0171)  acc1: 71.8750 (74.2577)  acc5: 90.6250 (93.9808)  time: 0.1308  data: 0.0002  max mem: 17867
[09:37:25.762788] Test:  [1000/1563]  eta: 0:01:13  loss: 1.2524 (1.1590)  acc1: 68.7500 (71.6409)  acc5: 90.6250 (91.5772)  time: 0.1299  data: 0.0002  max mem: 17867
[09:38:30.781380] Test:  [1500/1563]  eta: 0:00:08  loss: 0.7605 (1.2390)  acc1: 81.2500 (69.8305)  acc5: 93.7500 (90.3481)  time: 0.1312  data: 0.0002  max mem: 17867
[09:38:38.836427] Test:  [1562/1563]  eta: 0:00:00  loss: 0.6770 (1.2395)  acc1: 84.3750 (69.8200)  acc5: 96.8750 (90.3300)  time: 0.1265  data: 0.0001  max mem: 17867
[09:38:38.913275] Test: Total time: 0:03:24 (0.1308 s / it)
[09:38:38.915720] * Acc@1 69.820 Acc@5 90.330 loss 1.239
[09:38:38.915898] Accuracy of the network on the 50000 test images: 69.8%
[09:38:38.915919] Max accuracy: 69.82%
[09:38:38.922596] log_dir: ./output_dir_cml_spikformer
[09:38:40.540643] Epoch: [86]  [   0/5004]  eta: 2:14:53  lr: 0.000758  loss: 2.9756 (2.9756)  time: 1.6174  data: 1.1808  max mem: 17867
[09:50:20.453881] Epoch: [86]  [2000/5004]  eta: 0:17:33  lr: 0.000754  loss: 2.8078 (3.0604)  time: 0.3466  data: 0.0002  max mem: 17867
[10:02:00.842618] Epoch: [86]  [4000/5004]  eta: 0:05:51  lr: 0.000751  loss: 3.0124 (3.0621)  time: 0.3469  data: 0.0002  max mem: 17867
[10:07:52.980687] Epoch: [86]  [5003/5004]  eta: 0:00:00  lr: 0.000749  loss: 3.0401 (3.0661)  time: 0.3460  data: 0.0007  max mem: 17867
[10:07:53.353063] Epoch: [86] Total time: 0:29:14 (0.3506 s / it)
[10:07:53.358041] Averaged stats: lr: 0.000749  loss: 3.0401 (3.0629)
[10:07:54.706338] Test:  [   0/1563]  eta: 0:35:01  loss: 0.4649 (0.4649)  acc1: 90.6250 (90.6250)  acc5: 96.8750 (96.8750)  time: 1.3447  data: 1.2060  max mem: 17867
[10:08:59.694981] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.7597 (0.9761)  acc1: 71.8750 (74.9813)  acc5: 93.7500 (94.3862)  time: 0.1300  data: 0.0002  max mem: 17867
[10:10:04.734266] Test:  [1000/1563]  eta: 0:01:13  loss: 1.5662 (1.1280)  acc1: 59.3750 (72.1248)  acc5: 90.6250 (92.0642)  time: 0.1303  data: 0.0002  max mem: 17867
[10:11:09.733735] Test:  [1500/1563]  eta: 0:00:08  loss: 0.7151 (1.2295)  acc1: 84.3750 (70.0491)  acc5: 93.7500 (90.6812)  time: 0.1299  data: 0.0002  max mem: 17867
[10:11:17.717788] Test:  [1562/1563]  eta: 0:00:00  loss: 0.6792 (1.2311)  acc1: 87.5000 (70.0220)  acc5: 96.8750 (90.6700)  time: 0.1262  data: 0.0001  max mem: 17867
[10:11:17.788681] Test: Total time: 0:03:24 (0.1308 s / it)
[10:11:18.045850] * Acc@1 70.022 Acc@5 90.670 loss 1.231
[10:11:18.046037] Accuracy of the network on the 50000 test images: 70.0%
[10:11:18.046059] Max accuracy: 70.02%
[10:11:18.088099] log_dir: ./output_dir_cml_spikformer
[10:11:19.551731] Epoch: [87]  [   0/5004]  eta: 2:02:00  lr: 0.000749  loss: 2.4698 (2.4698)  time: 1.4629  data: 1.1307  max mem: 17867
[10:23:00.261950] Epoch: [87]  [2000/5004]  eta: 0:17:34  lr: 0.000745  loss: 2.8892 (3.0330)  time: 0.3495  data: 0.0002  max mem: 17867
[10:34:40.474804] Epoch: [87]  [4000/5004]  eta: 0:05:51  lr: 0.000741  loss: 2.8245 (3.0421)  time: 0.3470  data: 0.0002  max mem: 17867
[10:40:31.376819] Epoch: [87]  [5003/5004]  eta: 0:00:00  lr: 0.000739  loss: 3.0905 (3.0456)  time: 0.3437  data: 0.0006  max mem: 17867
[10:40:31.770086] Epoch: [87] Total time: 0:29:13 (0.3505 s / it)
[10:40:31.772414] Averaged stats: lr: 0.000739  loss: 3.0905 (3.0574)
[10:40:32.797173] Test:  [   0/1563]  eta: 0:26:36  loss: 0.3142 (0.3142)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.0214  data: 0.8774  max mem: 17867
[10:41:37.849403] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.0259 (0.9863)  acc1: 68.7500 (75.3493)  acc5: 93.7500 (94.4424)  time: 0.1301  data: 0.0002  max mem: 17867
[10:42:42.838963] Test:  [1000/1563]  eta: 0:01:13  loss: 1.3751 (1.1675)  acc1: 65.6250 (71.5909)  acc5: 90.6250 (91.6771)  time: 0.1298  data: 0.0002  max mem: 17867
[10:43:47.831207] Test:  [1500/1563]  eta: 0:00:08  loss: 0.7353 (1.2557)  acc1: 84.3750 (69.7306)  acc5: 96.8750 (90.3085)  time: 0.1298  data: 0.0002  max mem: 17867
[10:43:55.823655] Test:  [1562/1563]  eta: 0:00:00  loss: 0.7049 (1.2571)  acc1: 81.2500 (69.6780)  acc5: 96.8750 (90.2680)  time: 0.1266  data: 0.0001  max mem: 17867
[10:43:55.900387] Test: Total time: 0:03:24 (0.1306 s / it)
[10:43:56.096093] * Acc@1 69.678 Acc@5 90.268 loss 1.257
[10:43:56.096271] Accuracy of the network on the 50000 test images: 69.7%
[10:43:56.096303] Max accuracy: 70.02%
[10:43:56.104459] log_dir: ./output_dir_cml_spikformer
[10:43:57.605974] Epoch: [88]  [   0/5004]  eta: 2:05:07  lr: 0.000739  loss: 2.9727 (2.9727)  time: 1.5002  data: 1.1705  max mem: 17867
[10:55:37.801473] Epoch: [88]  [2000/5004]  eta: 0:17:33  lr: 0.000736  loss: 3.0363 (3.0439)  time: 0.3529  data: 0.0002  max mem: 17867
[11:07:17.736173] Epoch: [88]  [4000/5004]  eta: 0:05:51  lr: 0.000732  loss: 3.0791 (3.0441)  time: 0.3446  data: 0.0002  max mem: 17867
[11:13:08.727182] Epoch: [88]  [5003/5004]  eta: 0:00:00  lr: 0.000730  loss: 2.9241 (3.0460)  time: 0.3489  data: 0.0011  max mem: 17867
[11:13:09.102623] Epoch: [88] Total time: 0:29:12 (0.3503 s / it)
[11:13:09.107089] Averaged stats: lr: 0.000730  loss: 2.9241 (3.0519)
[11:13:10.408710] Test:  [   0/1563]  eta: 0:33:46  loss: 0.4323 (0.4323)  acc1: 87.5000 (87.5000)  acc5: 96.8750 (96.8750)  time: 1.2965  data: 1.1597  max mem: 17867
[11:14:15.452917] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.9812 (1.0159)  acc1: 71.8750 (74.4573)  acc5: 93.7500 (93.9059)  time: 0.1299  data: 0.0002  max mem: 17867
[11:15:20.464496] Test:  [1000/1563]  eta: 0:01:13  loss: 1.1583 (1.1659)  acc1: 71.8750 (71.5659)  acc5: 93.7500 (91.6740)  time: 0.1300  data: 0.0002  max mem: 17867
[11:16:25.466293] Test:  [1500/1563]  eta: 0:00:08  loss: 0.7957 (1.2506)  acc1: 81.2500 (69.7972)  acc5: 93.7500 (90.4480)  time: 0.1299  data: 0.0002  max mem: 17867
[11:16:33.460627] Test:  [1562/1563]  eta: 0:00:00  loss: 0.6623 (1.2497)  acc1: 87.5000 (69.7940)  acc5: 96.8750 (90.4760)  time: 0.1264  data: 0.0001  max mem: 17867
[11:16:33.582542] Test: Total time: 0:03:24 (0.1308 s / it)
[11:16:33.612459] * Acc@1 69.794 Acc@5 90.476 loss 1.250
[11:16:33.612588] Accuracy of the network on the 50000 test images: 69.8%
[11:16:33.612610] Max accuracy: 70.02%
[11:16:33.631484] log_dir: ./output_dir_cml_spikformer
[11:16:35.154358] Epoch: [89]  [   0/5004]  eta: 2:06:55  lr: 0.000730  loss: 2.8939 (2.8939)  time: 1.5220  data: 1.0237  max mem: 17867
[11:28:13.372196] Epoch: [89]  [2000/5004]  eta: 0:17:30  lr: 0.000726  loss: 2.9424 (3.0495)  time: 0.3460  data: 0.0002  max mem: 17867
[11:39:51.487728] Epoch: [89]  [4000/5004]  eta: 0:05:50  lr: 0.000722  loss: 2.9299 (3.0510)  time: 0.3551  data: 0.0002  max mem: 17867
[11:45:41.672910] Epoch: [89]  [5003/5004]  eta: 0:00:00  lr: 0.000720  loss: 3.0429 (3.0532)  time: 0.3436  data: 0.0011  max mem: 17867
[11:45:42.032507] Epoch: [89] Total time: 0:29:08 (0.3494 s / it)
[11:45:42.045187] Averaged stats: lr: 0.000720  loss: 3.0429 (3.0458)
[11:45:42.991942] Test:  [   0/1563]  eta: 0:24:34  loss: 0.3862 (0.3862)  acc1: 90.6250 (90.6250)  acc5: 96.8750 (96.8750)  time: 0.9434  data: 0.7991  max mem: 17867
[11:46:47.985318] Test:  [ 500/1563]  eta: 0:02:19  loss: 0.9775 (1.0274)  acc1: 65.6250 (74.4261)  acc5: 93.7500 (94.0619)  time: 0.1299  data: 0.0002  max mem: 17867
[11:47:53.036436] Test:  [1000/1563]  eta: 0:01:13  loss: 1.3842 (1.1782)  acc1: 62.5000 (71.2100)  acc5: 90.6250 (91.6178)  time: 0.1299  data: 0.0002  max mem: 17867
[11:48:58.031850] Test:  [1500/1563]  eta: 0:00:08  loss: 0.7722 (1.2859)  acc1: 84.3750 (69.0560)  acc5: 93.7500 (89.9754)  time: 0.1299  data: 0.0002  max mem: 17867
[11:49:06.075290] Test:  [1562/1563]  eta: 0:00:00  loss: 0.5302 (1.2810)  acc1: 90.6250 (69.1100)  acc5: 96.8750 (90.0620)  time: 0.1293  data: 0.0001  max mem: 17867
[11:49:06.147863] Test: Total time: 0:03:24 (0.1306 s / it)
[11:49:06.382960] * Acc@1 69.110 Acc@5 90.062 loss 1.281
[11:49:06.383134] Accuracy of the network on the 50000 test images: 69.1%
[11:49:06.383156] Max accuracy: 70.02%
[11:49:06.389878] log_dir: ./output_dir_cml_spikformer
[11:49:07.867454] Epoch: [90]  [   0/5004]  eta: 2:03:06  lr: 0.000720  loss: 3.2370 (3.2370)  time: 1.4762  data: 0.9767  max mem: 17867
[12:00:47.354207] Epoch: [90]  [2000/5004]  eta: 0:17:32  lr: 0.000717  loss: 3.0448 (3.0315)  time: 0.3619  data: 0.0002  max mem: 17867
[12:12:25.927532] Epoch: [90]  [4000/5004]  eta: 0:05:51  lr: 0.000713  loss: 2.9752 (3.0354)  time: 0.3448  data: 0.0002  max mem: 17867
[12:18:16.190605] Epoch: [90]  [5003/5004]  eta: 0:00:00  lr: 0.000711  loss: 2.9675 (3.0390)  time: 0.3488  data: 0.0011  max mem: 17867
[12:18:16.564208] Epoch: [90] Total time: 0:29:10 (0.3498 s / it)
[12:18:16.570417] Averaged stats: lr: 0.000711  loss: 2.9675 (3.0391)
[12:18:17.737474] Test:  [   0/1563]  eta: 0:30:18  loss: 0.2696 (0.2696)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.1635  data: 1.0226  max mem: 17867
[12:19:22.848424] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.9672 (0.9587)  acc1: 75.0000 (74.8503)  acc5: 96.8750 (94.3426)  time: 0.1300  data: 0.0002  max mem: 17867
[12:20:27.906167] Test:  [1000/1563]  eta: 0:01:13  loss: 1.3517 (1.1128)  acc1: 68.7500 (72.0779)  acc5: 90.6250 (92.0673)  time: 0.1311  data: 0.0002  max mem: 17867
[12:21:32.883867] Test:  [1500/1563]  eta: 0:00:08  loss: 0.6249 (1.2251)  acc1: 87.5000 (69.8451)  acc5: 93.7500 (90.3627)  time: 0.1299  data: 0.0002  max mem: 17867
[12:21:40.866476] Test:  [1562/1563]  eta: 0:00:00  loss: 0.8725 (1.2251)  acc1: 81.2500 (69.8100)  acc5: 93.7500 (90.3820)  time: 0.1262  data: 0.0001  max mem: 17867
[12:21:40.928900] Test: Total time: 0:03:24 (0.1307 s / it)
[12:21:41.115940] * Acc@1 69.810 Acc@5 90.382 loss 1.225
[12:21:41.116077] Accuracy of the network on the 50000 test images: 69.8%
[12:21:41.116099] Max accuracy: 70.02%
[12:21:41.145649] log_dir: ./output_dir_cml_spikformer
[12:21:42.634651] Epoch: [91]  [   0/5004]  eta: 2:04:06  lr: 0.000711  loss: 3.1295 (3.1295)  time: 1.4882  data: 1.1533  max mem: 17867
[12:33:20.856909] Epoch: [91]  [2000/5004]  eta: 0:17:30  lr: 0.000707  loss: 2.9908 (3.0255)  time: 0.3450  data: 0.0002  max mem: 17867
[12:44:58.067963] Epoch: [91]  [4000/5004]  eta: 0:05:50  lr: 0.000703  loss: 3.0283 (3.0286)  time: 0.3494  data: 0.0002  max mem: 17867
[12:50:47.800867] Epoch: [91]  [5003/5004]  eta: 0:00:00  lr: 0.000701  loss: 3.0188 (3.0345)  time: 0.3441  data: 0.0011  max mem: 17867
[12:50:48.184186] Epoch: [91] Total time: 0:29:07 (0.3491 s / it)
[12:50:48.185227] Averaged stats: lr: 0.000701  loss: 3.0188 (3.0354)
[12:50:49.247727] Test:  [   0/1563]  eta: 0:27:34  loss: 0.6738 (0.6738)  acc1: 87.5000 (87.5000)  acc5: 96.8750 (96.8750)  time: 1.0586  data: 0.8942  max mem: 17867
[12:51:54.319204] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.1778 (0.9702)  acc1: 65.6250 (74.6382)  acc5: 96.8750 (94.3862)  time: 0.1299  data: 0.0002  max mem: 17867
[12:52:59.303977] Test:  [1000/1563]  eta: 0:01:13  loss: 1.5035 (1.1380)  acc1: 62.5000 (71.5972)  acc5: 87.5000 (91.9705)  time: 0.1299  data: 0.0002  max mem: 17867
[12:54:04.283719] Test:  [1500/1563]  eta: 0:00:08  loss: 0.6634 (1.2312)  acc1: 84.3750 (69.8160)  acc5: 96.8750 (90.6708)  time: 0.1299  data: 0.0002  max mem: 17867
[12:54:12.274942] Test:  [1562/1563]  eta: 0:00:00  loss: 0.5465 (1.2282)  acc1: 90.6250 (69.9080)  acc5: 96.8750 (90.7360)  time: 0.1262  data: 0.0001  max mem: 17867
[12:54:12.350183] Test: Total time: 0:03:24 (0.1306 s / it)
[12:54:12.423592] * Acc@1 69.908 Acc@5 90.736 loss 1.228
[12:54:12.423762] Accuracy of the network on the 50000 test images: 69.9%
[12:54:12.423784] Max accuracy: 70.02%
[12:54:12.430579] log_dir: ./output_dir_cml_spikformer
[12:54:13.982828] Epoch: [92]  [   0/5004]  eta: 2:09:21  lr: 0.000701  loss: 2.7667 (2.7667)  time: 1.5511  data: 1.0756  max mem: 17867
[13:05:54.752174] Epoch: [92]  [2000/5004]  eta: 0:17:34  lr: 0.000698  loss: 2.9338 (3.0268)  time: 0.3571  data: 0.0002  max mem: 17867
[13:17:32.716392] Epoch: [92]  [4000/5004]  eta: 0:05:51  lr: 0.000694  loss: 3.0386 (3.0397)  time: 0.3483  data: 0.0002  max mem: 17867
[13:23:22.526119] Epoch: [92]  [5003/5004]  eta: 0:00:00  lr: 0.000692  loss: 3.0255 (3.0358)  time: 0.3437  data: 0.0012  max mem: 17867
[13:23:22.919599] Epoch: [92] Total time: 0:29:10 (0.3498 s / it)
[13:23:22.924444] Averaged stats: lr: 0.000692  loss: 3.0255 (3.0320)
[13:23:24.033393] Test:  [   0/1563]  eta: 0:28:44  loss: 0.3761 (0.3761)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.1030  data: 0.9461  max mem: 17867
[13:24:29.058377] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.9158 (0.9815)  acc1: 68.7500 (74.7131)  acc5: 96.8750 (94.0245)  time: 0.1299  data: 0.0002  max mem: 17867
[13:25:34.121179] Test:  [1000/1563]  eta: 0:01:13  loss: 1.6603 (1.1420)  acc1: 53.1250 (71.8219)  acc5: 90.6250 (91.6927)  time: 0.1300  data: 0.0002  max mem: 17867
[13:26:39.257382] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5597 (1.2202)  acc1: 87.5000 (69.9971)  acc5: 96.8750 (90.5438)  time: 0.1299  data: 0.0002  max mem: 17867
[13:26:47.243132] Test:  [1562/1563]  eta: 0:00:00  loss: 0.5700 (1.2225)  acc1: 84.3750 (69.9640)  acc5: 96.8750 (90.5200)  time: 0.1262  data: 0.0001  max mem: 17867
[13:26:47.326546] Test: Total time: 0:03:24 (0.1308 s / it)
[13:26:47.577430] * Acc@1 69.964 Acc@5 90.520 loss 1.222
[13:26:47.577590] Accuracy of the network on the 50000 test images: 70.0%
[13:26:47.577615] Max accuracy: 70.02%
[13:26:47.622504] log_dir: ./output_dir_cml_spikformer
[13:26:49.137460] Epoch: [93]  [   0/5004]  eta: 2:06:17  lr: 0.000692  loss: 2.8886 (2.8886)  time: 1.5142  data: 1.1423  max mem: 17867
[13:38:28.545892] Epoch: [93]  [2000/5004]  eta: 0:17:32  lr: 0.000688  loss: 3.0349 (3.0225)  time: 0.3479  data: 0.0002  max mem: 17867
[13:50:07.349640] Epoch: [93]  [4000/5004]  eta: 0:05:51  lr: 0.000684  loss: 3.1046 (3.0228)  time: 0.3480  data: 0.0002  max mem: 17867
[13:55:58.326949] Epoch: [93]  [5003/5004]  eta: 0:00:00  lr: 0.000682  loss: 3.0444 (3.0257)  time: 0.3521  data: 0.0012  max mem: 17867
[13:55:58.703856] Epoch: [93] Total time: 0:29:11 (0.3499 s / it)
[13:55:58.704549] Averaged stats: lr: 0.000682  loss: 3.0444 (3.0250)
[13:55:59.914366] Test:  [   0/1563]  eta: 0:31:24  loss: 0.2941 (0.2941)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.2059  data: 1.0696  max mem: 17867
[13:57:05.074654] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.0021 (0.9993)  acc1: 71.8750 (74.6632)  acc5: 93.7500 (94.2552)  time: 0.1308  data: 0.0002  max mem: 17867
[13:58:10.089056] Test:  [1000/1563]  eta: 0:01:13  loss: 1.4640 (1.1529)  acc1: 56.2500 (71.6908)  acc5: 90.6250 (91.9393)  time: 0.1300  data: 0.0002  max mem: 17867
[13:59:15.368076] Test:  [1500/1563]  eta: 0:00:08  loss: 0.6080 (1.2422)  acc1: 87.5000 (69.8951)  acc5: 96.8750 (90.5063)  time: 0.1442  data: 0.0002  max mem: 17867
[13:59:23.414856] Test:  [1562/1563]  eta: 0:00:00  loss: 0.6342 (1.2469)  acc1: 87.5000 (69.7800)  acc5: 93.7500 (90.4620)  time: 0.1272  data: 0.0001  max mem: 17867
[13:59:23.503824] Test: Total time: 0:03:24 (0.1310 s / it)
[13:59:23.504919] * Acc@1 69.780 Acc@5 90.462 loss 1.247
[13:59:23.505029] Accuracy of the network on the 50000 test images: 69.8%
[13:59:23.505048] Max accuracy: 70.02%
[13:59:23.511862] log_dir: ./output_dir_cml_spikformer
[13:59:25.039925] Epoch: [94]  [   0/5004]  eta: 2:07:20  lr: 0.000682  loss: 3.2246 (3.2246)  time: 1.5269  data: 0.9988  max mem: 17867
[14:11:05.931868] Epoch: [94]  [2000/5004]  eta: 0:17:34  lr: 0.000679  loss: 3.0168 (3.0074)  time: 0.3478  data: 0.0002  max mem: 17867
[14:22:46.730389] Epoch: [94]  [4000/5004]  eta: 0:05:52  lr: 0.000675  loss: 2.8440 (3.0172)  time: 0.3518  data: 0.0002  max mem: 17867
[14:28:37.380773] Epoch: [94]  [5003/5004]  eta: 0:00:00  lr: 0.000673  loss: 3.0762 (3.0191)  time: 0.3452  data: 0.0011  max mem: 17867
[14:28:37.759097] Epoch: [94] Total time: 0:29:14 (0.3506 s / it)
[14:28:37.759809] Averaged stats: lr: 0.000673  loss: 3.0762 (3.0200)
[14:28:38.878305] Test:  [   0/1563]  eta: 0:29:02  loss: 0.3524 (0.3524)  acc1: 96.8750 (96.8750)  acc5: 100.0000 (100.0000)  time: 1.1149  data: 0.9760  max mem: 17867
[14:29:44.021049] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.9268 (0.9646)  acc1: 75.0000 (75.3368)  acc5: 96.8750 (94.3613)  time: 0.1300  data: 0.0002  max mem: 17867
[14:30:48.964800] Test:  [1000/1563]  eta: 0:01:13  loss: 1.1470 (1.1258)  acc1: 65.6250 (71.9187)  acc5: 90.6250 (91.9424)  time: 0.1305  data: 0.0002  max mem: 17867
[14:31:54.004719] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5973 (1.2161)  acc1: 84.3750 (70.0242)  acc5: 96.8750 (90.6375)  time: 0.1300  data: 0.0002  max mem: 17867
[14:32:02.015355] Test:  [1562/1563]  eta: 0:00:00  loss: 0.5658 (1.2166)  acc1: 87.5000 (70.0520)  acc5: 96.8750 (90.6220)  time: 0.1270  data: 0.0001  max mem: 17867
[14:32:02.095239] Test: Total time: 0:03:24 (0.1307 s / it)
[14:32:02.230293] * Acc@1 70.052 Acc@5 90.622 loss 1.217
[14:32:02.230452] Accuracy of the network on the 50000 test images: 70.1%
[14:32:02.230474] Max accuracy: 70.05%
[14:32:02.270082] log_dir: ./output_dir_cml_spikformer
[14:32:03.721549] Epoch: [95]  [   0/5004]  eta: 2:00:56  lr: 0.000673  loss: 2.9410 (2.9410)  time: 1.4502  data: 1.0325  max mem: 17867
[14:43:44.891310] Epoch: [95]  [2000/5004]  eta: 0:17:34  lr: 0.000669  loss: 3.2524 (3.0085)  time: 0.3541  data: 0.0002  max mem: 17867
[14:55:25.932335] Epoch: [95]  [4000/5004]  eta: 0:05:52  lr: 0.000665  loss: 2.9021 (3.0151)  time: 0.3446  data: 0.0002  max mem: 17867
[15:01:17.001625] Epoch: [95]  [5003/5004]  eta: 0:00:00  lr: 0.000663  loss: 3.0242 (3.0161)  time: 0.3488  data: 0.0012  max mem: 17867
[15:01:17.367161] Epoch: [95] Total time: 0:29:15 (0.3507 s / it)
[15:01:17.374675] Averaged stats: lr: 0.000663  loss: 3.0242 (3.0132)
[15:01:18.703914] Test:  [   0/1563]  eta: 0:34:32  loss: 0.3624 (0.3624)  acc1: 93.7500 (93.7500)  acc5: 100.0000 (100.0000)  time: 1.3257  data: 1.1869  max mem: 17867
[15:02:23.752424] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.8204 (0.9999)  acc1: 75.0000 (74.6944)  acc5: 93.7500 (94.1929)  time: 0.1300  data: 0.0002  max mem: 17867
[15:03:28.779187] Test:  [1000/1563]  eta: 0:01:13  loss: 1.3724 (1.1356)  acc1: 68.7500 (72.0748)  acc5: 93.7500 (92.0611)  time: 0.1299  data: 0.0002  max mem: 17867
[15:04:33.771905] Test:  [1500/1563]  eta: 0:00:08  loss: 0.8099 (1.2217)  acc1: 78.1250 (70.2698)  acc5: 93.7500 (90.8165)  time: 0.1299  data: 0.0002  max mem: 17867
[15:04:41.760080] Test:  [1562/1563]  eta: 0:00:00  loss: 0.5321 (1.2272)  acc1: 87.5000 (70.1580)  acc5: 96.8750 (90.7620)  time: 0.1262  data: 0.0001  max mem: 17867
[15:04:41.823599] Test: Total time: 0:03:24 (0.1308 s / it)
[15:04:42.046797] * Acc@1 70.158 Acc@5 90.762 loss 1.227
[15:04:42.046966] Accuracy of the network on the 50000 test images: 70.2%
[15:04:42.046988] Max accuracy: 70.16%
[15:04:42.078394] log_dir: ./output_dir_cml_spikformer
[15:04:43.555557] Epoch: [96]  [   0/5004]  eta: 2:03:05  lr: 0.000663  loss: 3.2070 (3.2070)  time: 1.4758  data: 1.0145  max mem: 17867
[15:16:24.660201] Epoch: [96]  [2000/5004]  eta: 0:17:34  lr: 0.000659  loss: 2.9152 (2.9913)  time: 0.3451  data: 0.0002  max mem: 17867
[15:28:05.617101] Epoch: [96]  [4000/5004]  eta: 0:05:52  lr: 0.000655  loss: 3.0555 (2.9997)  time: 0.3511  data: 0.0002  max mem: 17867
[15:33:56.806505] Epoch: [96]  [5003/5004]  eta: 0:00:00  lr: 0.000654  loss: 2.8766 (3.0051)  time: 0.3435  data: 0.0011  max mem: 17867
[15:33:57.174396] Epoch: [96] Total time: 0:29:15 (0.3507 s / it)
[15:33:57.181490] Averaged stats: lr: 0.000654  loss: 2.8766 (3.0113)
[15:33:58.206135] Test:  [   0/1563]  eta: 0:26:35  loss: 0.8917 (0.8917)  acc1: 81.2500 (81.2500)  acc5: 96.8750 (96.8750)  time: 1.0210  data: 0.8369  max mem: 17867
[15:35:03.325745] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.9909 (0.9649)  acc1: 75.0000 (74.4636)  acc5: 93.7500 (94.3987)  time: 0.1299  data: 0.0002  max mem: 17867
[15:36:08.326497] Test:  [1000/1563]  eta: 0:01:13  loss: 1.4970 (1.1148)  acc1: 59.3750 (71.9811)  acc5: 90.6250 (92.0704)  time: 0.1300  data: 0.0002  max mem: 17867
[15:37:13.391798] Test:  [1500/1563]  eta: 0:00:08  loss: 0.6406 (1.1951)  acc1: 84.3750 (70.4947)  acc5: 93.7500 (90.8519)  time: 0.1299  data: 0.0002  max mem: 17867
[15:37:21.379759] Test:  [1562/1563]  eta: 0:00:00  loss: 0.5886 (1.1970)  acc1: 87.5000 (70.4460)  acc5: 96.8750 (90.8500)  time: 0.1262  data: 0.0001  max mem: 17867
[15:37:21.456836] Test: Total time: 0:03:24 (0.1307 s / it)
[15:37:21.486298] * Acc@1 70.446 Acc@5 90.850 loss 1.197
[15:37:21.486434] Accuracy of the network on the 50000 test images: 70.4%
[15:37:21.486453] Max accuracy: 70.45%
[15:37:21.520695] log_dir: ./output_dir_cml_spikformer
[15:37:23.125824] Epoch: [97]  [   0/5004]  eta: 2:13:45  lr: 0.000654  loss: 2.8736 (2.8736)  time: 1.6038  data: 1.1022  max mem: 17867
[15:49:03.482480] Epoch: [97]  [2000/5004]  eta: 0:17:33  lr: 0.000650  loss: 2.9871 (3.0026)  time: 0.3553  data: 0.0002  max mem: 17867
[16:00:43.027144] Epoch: [97]  [4000/5004]  eta: 0:05:51  lr: 0.000646  loss: 3.0226 (2.9987)  time: 0.3446  data: 0.0002  max mem: 17867
[16:06:34.722740] Epoch: [97]  [5003/5004]  eta: 0:00:00  lr: 0.000644  loss: 3.0104 (2.9989)  time: 0.3458  data: 0.0011  max mem: 17867
[16:06:35.081053] Epoch: [97] Total time: 0:29:13 (0.3504 s / it)
[16:06:35.081836] Averaged stats: lr: 0.000644  loss: 3.0104 (3.0021)
[16:06:36.148671] Test:  [   0/1563]  eta: 0:27:41  loss: 0.4861 (0.4861)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.0633  data: 0.9254  max mem: 17867
[16:07:41.147694] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.9859 (0.9935)  acc1: 75.0000 (75.7298)  acc5: 93.7500 (94.4174)  time: 0.1298  data: 0.0002  max mem: 17867
[16:08:46.118302] Test:  [1000/1563]  eta: 0:01:13  loss: 1.3555 (1.1373)  acc1: 62.5000 (72.8178)  acc5: 90.6250 (92.1828)  time: 0.1299  data: 0.0002  max mem: 17867
[16:09:51.132981] Test:  [1500/1563]  eta: 0:00:08  loss: 0.6570 (1.2309)  acc1: 87.5000 (70.6737)  acc5: 93.7500 (90.7312)  time: 0.1301  data: 0.0002  max mem: 17867
[16:09:59.111733] Test:  [1562/1563]  eta: 0:00:00  loss: 0.5849 (1.2290)  acc1: 87.5000 (70.6460)  acc5: 96.8750 (90.7580)  time: 0.1262  data: 0.0001  max mem: 17867
[16:09:59.169630] Test: Total time: 0:03:24 (0.1306 s / it)
[16:09:59.387193] * Acc@1 70.646 Acc@5 90.758 loss 1.229
[16:09:59.387366] Accuracy of the network on the 50000 test images: 70.6%
[16:09:59.387387] Max accuracy: 70.65%
[16:09:59.453755] log_dir: ./output_dir_cml_spikformer
[16:10:00.903700] Epoch: [98]  [   0/5004]  eta: 2:00:48  lr: 0.000644  loss: 2.4632 (2.4632)  time: 1.4486  data: 0.8912  max mem: 17867
[16:21:40.710803] Epoch: [98]  [2000/5004]  eta: 0:17:32  lr: 0.000640  loss: 2.9012 (2.9939)  time: 0.3470  data: 0.0002  max mem: 17867
[16:33:20.129491] Epoch: [98]  [4000/5004]  eta: 0:05:51  lr: 0.000636  loss: 3.0654 (2.9970)  time: 0.3445  data: 0.0002  max mem: 17867
[16:39:10.837681] Epoch: [98]  [5003/5004]  eta: 0:00:00  lr: 0.000634  loss: 3.0770 (3.0022)  time: 0.3474  data: 0.0011  max mem: 17867
[16:39:11.211942] Epoch: [98] Total time: 0:29:11 (0.3501 s / it)
[16:39:11.237535] Averaged stats: lr: 0.000634  loss: 3.0770 (2.9983)
[16:39:12.568089] Test:  [   0/1563]  eta: 0:34:32  loss: 0.3643 (0.3643)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.3262  data: 1.0109  max mem: 17867
[16:40:17.816812] Test:  [ 500/1563]  eta: 0:02:21  loss: 1.0232 (0.9781)  acc1: 71.8750 (75.1559)  acc5: 93.7500 (94.2428)  time: 0.1308  data: 0.0002  max mem: 17867
[16:41:22.920452] Test:  [1000/1563]  eta: 0:01:14  loss: 1.5503 (1.1191)  acc1: 56.2500 (72.3807)  acc5: 87.5000 (92.2203)  time: 0.1299  data: 0.0002  max mem: 17867
[16:42:27.925526] Test:  [1500/1563]  eta: 0:00:08  loss: 0.7909 (1.2225)  acc1: 87.5000 (70.2573)  acc5: 96.8750 (90.7645)  time: 0.1300  data: 0.0002  max mem: 17867
[16:42:35.992024] Test:  [1562/1563]  eta: 0:00:00  loss: 0.6365 (1.2208)  acc1: 84.3750 (70.2740)  acc5: 96.8750 (90.7980)  time: 0.1303  data: 0.0001  max mem: 17867
[16:42:36.070762] Test: Total time: 0:03:24 (0.1310 s / it)
[16:42:36.072125] * Acc@1 70.274 Acc@5 90.798 loss 1.221
[16:42:36.072253] Accuracy of the network on the 50000 test images: 70.3%
[16:42:36.072282] Max accuracy: 70.65%
[16:42:36.098194] log_dir: ./output_dir_cml_spikformer
[16:42:37.534790] Epoch: [99]  [   0/5004]  eta: 1:59:42  lr: 0.000634  loss: 2.5014 (2.5014)  time: 1.4354  data: 0.9423  max mem: 17867
[16:54:16.884848] Epoch: [99]  [2000/5004]  eta: 0:17:32  lr: 0.000630  loss: 2.9761 (2.9792)  time: 0.3517  data: 0.0002  max mem: 17867
[17:05:55.553746] Epoch: [99]  [4000/5004]  eta: 0:05:51  lr: 0.000627  loss: 2.7786 (2.9878)  time: 0.3492  data: 0.0002  max mem: 17867
[17:11:45.963306] Epoch: [99]  [5003/5004]  eta: 0:00:00  lr: 0.000625  loss: 2.9878 (2.9870)  time: 0.3443  data: 0.0011  max mem: 17867
[17:11:46.347890] Epoch: [99] Total time: 0:29:10 (0.3498 s / it)
[17:11:46.352468] Averaged stats: lr: 0.000625  loss: 2.9878 (2.9902)
[17:11:47.464006] Test:  [   0/1563]  eta: 0:28:51  loss: 0.2991 (0.2991)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.1078  data: 0.9266  max mem: 17867
[17:12:52.439212] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.0868 (0.9497)  acc1: 75.0000 (75.8421)  acc5: 93.7500 (94.2677)  time: 0.1298  data: 0.0002  max mem: 17867
[17:13:57.380752] Test:  [1000/1563]  eta: 0:01:13  loss: 1.3429 (1.0840)  acc1: 65.6250 (73.1643)  acc5: 90.6250 (92.2858)  time: 0.1298  data: 0.0002  max mem: 17867
[17:15:02.367146] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5105 (1.1837)  acc1: 81.2500 (71.0547)  acc5: 96.8750 (90.9873)  time: 0.1303  data: 0.0002  max mem: 17867
[17:15:10.347180] Test:  [1562/1563]  eta: 0:00:00  loss: 0.5236 (1.1853)  acc1: 90.6250 (71.0560)  acc5: 96.8750 (90.9640)  time: 0.1262  data: 0.0001  max mem: 17867
[17:15:10.415178] Test: Total time: 0:03:24 (0.1306 s / it)
[17:15:10.957663] * Acc@1 71.056 Acc@5 90.964 loss 1.185
[17:15:10.957831] Accuracy of the network on the 50000 test images: 71.1%
[17:15:10.957862] Max accuracy: 71.06%
[17:15:11.111118] log_dir: ./output_dir_cml_spikformer
[17:15:12.607660] Epoch: [100]  [   0/5004]  eta: 2:04:44  lr: 0.000625  loss: 2.9153 (2.9153)  time: 1.4958  data: 1.0892  max mem: 17867
[17:26:51.635352] Epoch: [100]  [2000/5004]  eta: 0:17:31  lr: 0.000621  loss: 2.9107 (2.9719)  time: 0.3506  data: 0.0002  max mem: 17867
[17:38:30.754508] Epoch: [100]  [4000/5004]  eta: 0:05:51  lr: 0.000617  loss: 2.8811 (2.9850)  time: 0.3491  data: 0.0002  max mem: 17867
[17:44:21.344301] Epoch: [100]  [5003/5004]  eta: 0:00:00  lr: 0.000615  loss: 2.9953 (2.9851)  time: 0.3491  data: 0.0008  max mem: 17867
[17:44:21.705143] Epoch: [100] Total time: 0:29:10 (0.3498 s / it)
[17:44:21.713672] Averaged stats: lr: 0.000615  loss: 2.9953 (2.9841)
[17:44:22.763425] Test:  [   0/1563]  eta: 0:27:13  loss: 0.3358 (0.3358)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.0450  data: 0.8766  max mem: 17867
[17:45:27.784634] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.0448 (0.9517)  acc1: 75.0000 (76.2288)  acc5: 96.8750 (94.5546)  time: 0.1299  data: 0.0002  max mem: 17867
[17:46:32.774951] Test:  [1000/1563]  eta: 0:01:13  loss: 1.2569 (1.0961)  acc1: 68.7500 (73.1300)  acc5: 93.7500 (92.4419)  time: 0.1299  data: 0.0002  max mem: 17867
[17:47:37.786675] Test:  [1500/1563]  eta: 0:00:08  loss: 0.6747 (1.1913)  acc1: 81.2500 (71.0818)  acc5: 96.8750 (91.1018)  time: 0.1299  data: 0.0002  max mem: 17867
[17:47:45.769172] Test:  [1562/1563]  eta: 0:00:00  loss: 0.6057 (1.1889)  acc1: 84.3750 (71.1280)  acc5: 93.7500 (91.1380)  time: 0.1262  data: 0.0001  max mem: 17867
[17:47:45.842970] Test: Total time: 0:03:24 (0.1306 s / it)
[17:47:45.996925] * Acc@1 71.128 Acc@5 91.138 loss 1.189
[17:47:45.997065] Accuracy of the network on the 50000 test images: 71.1%
[17:47:45.997086] Max accuracy: 71.13%
[17:47:46.014067] log_dir: ./output_dir_cml_spikformer
[17:47:47.500129] Epoch: [101]  [   0/5004]  eta: 2:03:50  lr: 0.000615  loss: 2.8321 (2.8321)  time: 1.4848  data: 1.0954  max mem: 17867
[17:59:28.845414] Epoch: [101]  [2000/5004]  eta: 0:17:35  lr: 0.000611  loss: 3.0843 (2.9589)  time: 0.3485  data: 0.0003  max mem: 17867
[18:11:08.700636] Epoch: [101]  [4000/5004]  eta: 0:05:51  lr: 0.000607  loss: 2.9690 (2.9757)  time: 0.3501  data: 0.0002  max mem: 17867
[18:16:59.942268] Epoch: [101]  [5003/5004]  eta: 0:00:00  lr: 0.000605  loss: 2.8318 (2.9739)  time: 0.3440  data: 0.0013  max mem: 17867
[18:17:00.307468] Epoch: [101] Total time: 0:29:14 (0.3506 s / it)
[18:17:00.316488] Averaged stats: lr: 0.000605  loss: 2.8318 (2.9801)
[18:17:01.353153] Test:  [   0/1563]  eta: 0:26:54  loss: 0.4737 (0.4737)  acc1: 90.6250 (90.6250)  acc5: 96.8750 (96.8750)  time: 1.0331  data: 0.8915  max mem: 17867
[18:18:06.340818] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.7749 (0.9692)  acc1: 78.1250 (75.9356)  acc5: 93.7500 (94.5796)  time: 0.1301  data: 0.0002  max mem: 17867
[18:19:11.301539] Test:  [1000/1563]  eta: 0:01:13  loss: 1.7328 (1.1235)  acc1: 56.2500 (72.6086)  acc5: 87.5000 (92.2609)  time: 0.1298  data: 0.0002  max mem: 17867
[18:20:16.262288] Test:  [1500/1563]  eta: 0:00:08  loss: 0.6920 (1.1975)  acc1: 84.3750 (70.9360)  acc5: 96.8750 (91.1642)  time: 0.1298  data: 0.0002  max mem: 17867
[18:20:24.239465] Test:  [1562/1563]  eta: 0:00:00  loss: 0.5310 (1.1982)  acc1: 90.6250 (70.9020)  acc5: 96.8750 (91.1480)  time: 0.1262  data: 0.0001  max mem: 17867
[18:20:24.320625] Test: Total time: 0:03:24 (0.1305 s / it)
[18:20:24.588054] * Acc@1 70.902 Acc@5 91.148 loss 1.198
[18:20:24.588234] Accuracy of the network on the 50000 test images: 70.9%
[18:20:24.588271] Max accuracy: 71.13%
[18:20:24.612037] log_dir: ./output_dir_cml_spikformer
[18:20:26.034280] Epoch: [102]  [   0/5004]  eta: 1:58:28  lr: 0.000605  loss: 2.6459 (2.6459)  time: 1.4206  data: 0.9708  max mem: 17867
[18:32:09.609161] Epoch: [102]  [2000/5004]  eta: 0:17:38  lr: 0.000601  loss: 3.0247 (2.9479)  time: 0.3598  data: 0.0002  max mem: 17867
[18:43:49.903059] Epoch: [102]  [4000/5004]  eta: 0:05:52  lr: 0.000598  loss: 2.8707 (2.9640)  time: 0.3472  data: 0.0002  max mem: 17867
[18:49:41.499904] Epoch: [102]  [5003/5004]  eta: 0:00:00  lr: 0.000596  loss: 2.8316 (2.9651)  time: 0.3495  data: 0.0011  max mem: 17867
[18:49:41.874686] Epoch: [102] Total time: 0:29:17 (0.3512 s / it)
[18:49:41.875384] Averaged stats: lr: 0.000596  loss: 2.8316 (2.9736)
[18:49:42.950632] Test:  [   0/1563]  eta: 0:27:54  loss: 0.5289 (0.5289)  acc1: 90.6250 (90.6250)  acc5: 96.8750 (96.8750)  time: 1.0715  data: 0.9347  max mem: 17867
[18:50:48.046764] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.9468 (0.9361)  acc1: 78.1250 (76.1851)  acc5: 96.8750 (94.8977)  time: 0.1301  data: 0.0002  max mem: 17867
[18:51:52.957221] Test:  [1000/1563]  eta: 0:01:13  loss: 1.3708 (1.0786)  acc1: 62.5000 (73.2018)  acc5: 93.7500 (92.7010)  time: 0.1296  data: 0.0002  max mem: 17867
[18:52:57.974973] Test:  [1500/1563]  eta: 0:00:08  loss: 0.6946 (1.1677)  acc1: 84.3750 (71.3795)  acc5: 93.7500 (91.2662)  time: 0.1308  data: 0.0002  max mem: 17867
[18:53:06.032251] Test:  [1562/1563]  eta: 0:00:00  loss: 0.5167 (1.1658)  acc1: 90.6250 (71.4440)  acc5: 96.8750 (91.3040)  time: 0.1291  data: 0.0001  max mem: 17867
[18:53:06.148442] Test: Total time: 0:03:24 (0.1307 s / it)
[18:53:06.346894] * Acc@1 71.444 Acc@5 91.304 loss 1.166
[18:53:06.347034] Accuracy of the network on the 50000 test images: 71.4%
[18:53:06.347057] Max accuracy: 71.44%
[18:53:06.371858] log_dir: ./output_dir_cml_spikformer
[18:53:07.873123] Epoch: [103]  [   0/5004]  eta: 2:05:06  lr: 0.000596  loss: 2.1659 (2.1659)  time: 1.5002  data: 1.1441  max mem: 17867
[19:04:50.140359] Epoch: [103]  [2000/5004]  eta: 0:17:36  lr: 0.000592  loss: 2.7149 (2.9704)  time: 0.3446  data: 0.0002  max mem: 17867
[19:16:30.077201] Epoch: [103]  [4000/5004]  eta: 0:05:52  lr: 0.000588  loss: 2.9968 (2.9720)  time: 0.3445  data: 0.0002  max mem: 17867
[19:22:20.968727] Epoch: [103]  [5003/5004]  eta: 0:00:00  lr: 0.000586  loss: 2.8212 (2.9691)  time: 0.3445  data: 0.0011  max mem: 17867
[19:22:21.327728] Epoch: [103] Total time: 0:29:14 (0.3507 s / it)
[19:22:21.333038] Averaged stats: lr: 0.000586  loss: 2.8212 (2.9691)
[19:22:22.306288] Test:  [   0/1563]  eta: 0:25:15  loss: 0.3492 (0.3492)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 0.9695  data: 0.8264  max mem: 17867
[19:23:27.277261] Test:  [ 500/1563]  eta: 0:02:19  loss: 0.8851 (0.9499)  acc1: 78.1250 (75.9980)  acc5: 96.8750 (94.5047)  time: 0.1298  data: 0.0002  max mem: 17867
[19:24:32.229913] Test:  [1000/1563]  eta: 0:01:13  loss: 1.0786 (1.0879)  acc1: 68.7500 (73.1581)  acc5: 93.7500 (92.5200)  time: 0.1299  data: 0.0002  max mem: 17867
[19:25:37.235153] Test:  [1500/1563]  eta: 0:00:08  loss: 0.7166 (1.1754)  acc1: 84.3750 (71.3753)  acc5: 96.8750 (91.2225)  time: 0.1298  data: 0.0002  max mem: 17867
[19:25:45.217413] Test:  [1562/1563]  eta: 0:00:00  loss: 0.6385 (1.1763)  acc1: 87.5000 (71.3480)  acc5: 96.8750 (91.1880)  time: 0.1262  data: 0.0001  max mem: 17867
[19:25:45.283671] Test: Total time: 0:03:23 (0.1305 s / it)
[19:25:45.765384] * Acc@1 71.348 Acc@5 91.188 loss 1.176
[19:25:45.765628] Accuracy of the network on the 50000 test images: 71.3%
[19:25:45.765654] Max accuracy: 71.44%
[19:25:45.829950] log_dir: ./output_dir_cml_spikformer
[19:25:47.328407] Epoch: [104]  [   0/5004]  eta: 2:04:53  lr: 0.000586  loss: 2.8249 (2.8249)  time: 1.4976  data: 1.1184  max mem: 17867
[19:37:26.673389] Epoch: [104]  [2000/5004]  eta: 0:17:32  lr: 0.000582  loss: 2.9452 (2.9422)  time: 0.3508  data: 0.0003  max mem: 17867
[19:49:07.522438] Epoch: [104]  [4000/5004]  eta: 0:05:51  lr: 0.000578  loss: 3.0228 (2.9582)  time: 0.3526  data: 0.0002  max mem: 17867
[19:54:58.208569] Epoch: [104]  [5003/5004]  eta: 0:00:00  lr: 0.000576  loss: 2.9394 (2.9607)  time: 0.3430  data: 0.0011  max mem: 17867
[19:54:58.543879] Epoch: [104] Total time: 0:29:12 (0.3503 s / it)
[19:54:58.545838] Averaged stats: lr: 0.000576  loss: 2.9394 (2.9597)
[19:54:59.594360] Test:  [   0/1563]  eta: 0:27:11  loss: 0.4385 (0.4385)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.0439  data: 0.8938  max mem: 17867
[19:56:04.587475] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.0641 (0.9745)  acc1: 75.0000 (75.9294)  acc5: 93.7500 (94.5172)  time: 0.1299  data: 0.0002  max mem: 17867
[19:57:09.586935] Test:  [1000/1563]  eta: 0:01:13  loss: 1.3898 (1.1143)  acc1: 65.6250 (72.7554)  acc5: 93.7500 (92.4544)  time: 0.1299  data: 0.0002  max mem: 17867
[19:58:14.581901] Test:  [1500/1563]  eta: 0:00:08  loss: 0.6837 (1.1913)  acc1: 87.5000 (71.1900)  acc5: 96.8750 (91.3849)  time: 0.1299  data: 0.0002  max mem: 17867
[19:58:22.565266] Test:  [1562/1563]  eta: 0:00:00  loss: 0.6609 (1.1938)  acc1: 87.5000 (71.1300)  acc5: 96.8750 (91.3440)  time: 0.1262  data: 0.0001  max mem: 17867
[19:58:22.631202] Test: Total time: 0:03:24 (0.1306 s / it)
[19:58:22.914055] * Acc@1 71.130 Acc@5 91.344 loss 1.194
[19:58:22.914203] Accuracy of the network on the 50000 test images: 71.1%
[19:58:22.914224] Max accuracy: 71.44%
[19:58:22.949367] log_dir: ./output_dir_cml_spikformer
[19:58:24.550759] Epoch: [105]  [   0/5004]  eta: 2:13:29  lr: 0.000576  loss: 2.8482 (2.8482)  time: 1.6007  data: 1.1313  max mem: 17867
[20:10:04.247257] Epoch: [105]  [2000/5004]  eta: 0:17:32  lr: 0.000573  loss: 2.9516 (2.9541)  time: 0.3491  data: 0.0002  max mem: 17867
[20:21:44.149901] Epoch: [105]  [4000/5004]  eta: 0:05:51  lr: 0.000569  loss: 2.9623 (2.9623)  time: 0.3471  data: 0.0002  max mem: 17867
[20:27:34.590991] Epoch: [105]  [5003/5004]  eta: 0:00:00  lr: 0.000567  loss: 2.8060 (2.9599)  time: 0.3467  data: 0.0011  max mem: 17867
[20:27:34.934107] Epoch: [105] Total time: 0:29:11 (0.3501 s / it)
[20:27:34.935284] Averaged stats: lr: 0.000567  loss: 2.8060 (2.9565)
[20:27:35.963692] Test:  [   0/1563]  eta: 0:26:41  loss: 0.4483 (0.4483)  acc1: 90.6250 (90.6250)  acc5: 96.8750 (96.8750)  time: 1.0249  data: 0.8669  max mem: 17867
[20:28:40.958603] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.7803 (0.9001)  acc1: 75.0000 (76.4159)  acc5: 96.8750 (94.6544)  time: 0.1298  data: 0.0002  max mem: 17867
[20:29:45.903069] Test:  [1000/1563]  eta: 0:01:13  loss: 1.2814 (1.0369)  acc1: 65.6250 (73.7075)  acc5: 90.6250 (92.6105)  time: 0.1298  data: 0.0002  max mem: 17867
[20:30:50.877983] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5680 (1.1283)  acc1: 87.5000 (71.7771)  acc5: 93.7500 (91.3183)  time: 0.1300  data: 0.0002  max mem: 17867
[20:30:58.884958] Test:  [1562/1563]  eta: 0:00:00  loss: 0.5190 (1.1297)  acc1: 87.5000 (71.7380)  acc5: 96.8750 (91.3260)  time: 0.1262  data: 0.0001  max mem: 17867
[20:30:58.944330] Test: Total time: 0:03:24 (0.1305 s / it)
[20:30:59.140944] * Acc@1 71.738 Acc@5 91.326 loss 1.130
[20:30:59.141124] Accuracy of the network on the 50000 test images: 71.7%
[20:30:59.141146] Max accuracy: 71.74%
[20:30:59.174622] log_dir: ./output_dir_cml_spikformer
[20:31:00.654112] Epoch: [106]  [   0/5004]  eta: 2:03:19  lr: 0.000567  loss: 2.9093 (2.9093)  time: 1.4788  data: 1.1318  max mem: 17867
[20:42:41.239102] Epoch: [106]  [2000/5004]  eta: 0:17:33  lr: 0.000563  loss: 2.9076 (2.9620)  time: 0.3480  data: 0.0002  max mem: 17867
[20:54:22.107620] Epoch: [106]  [4000/5004]  eta: 0:05:52  lr: 0.000559  loss: 2.9461 (2.9556)  time: 0.3545  data: 0.0002  max mem: 17867
[21:00:12.845649] Epoch: [106]  [5003/5004]  eta: 0:00:00  lr: 0.000557  loss: 2.9527 (2.9534)  time: 0.3433  data: 0.0011  max mem: 17867
[21:00:13.204666] Epoch: [106] Total time: 0:29:14 (0.3505 s / it)
[21:00:13.205429] Averaged stats: lr: 0.000557  loss: 2.9527 (2.9502)
[21:00:14.240467] Test:  [   0/1563]  eta: 0:26:51  loss: 0.4901 (0.4901)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.0313  data: 0.8937  max mem: 17867
[21:01:19.257558] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.9105 (0.9359)  acc1: 75.0000 (75.9294)  acc5: 93.7500 (94.3862)  time: 0.1299  data: 0.0002  max mem: 17867
[21:02:24.300152] Test:  [1000/1563]  eta: 0:01:13  loss: 1.1233 (1.0702)  acc1: 78.1250 (73.1768)  acc5: 90.6250 (92.3420)  time: 0.1299  data: 0.0002  max mem: 17867
[21:03:29.302076] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5499 (1.1643)  acc1: 87.5000 (71.1130)  acc5: 96.8750 (90.9831)  time: 0.1299  data: 0.0002  max mem: 17867
[21:03:37.293940] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4499 (1.1662)  acc1: 87.5000 (71.0480)  acc5: 96.8750 (90.9780)  time: 0.1263  data: 0.0001  max mem: 17867
[21:03:37.371694] Test: Total time: 0:03:24 (0.1306 s / it)
[21:03:37.469351] * Acc@1 71.048 Acc@5 90.978 loss 1.166
[21:03:37.469510] Accuracy of the network on the 50000 test images: 71.0%
[21:03:37.469534] Max accuracy: 71.74%
[21:03:37.478079] log_dir: ./output_dir_cml_spikformer
[21:03:38.994294] Epoch: [107]  [   0/5004]  eta: 2:06:20  lr: 0.000557  loss: 3.2464 (3.2464)  time: 1.5149  data: 1.0480  max mem: 17867
[21:15:19.359062] Epoch: [107]  [2000/5004]  eta: 0:17:33  lr: 0.000553  loss: 2.8765 (2.9356)  time: 0.3538  data: 0.0002  max mem: 17867
[21:27:00.737123] Epoch: [107]  [4000/5004]  eta: 0:05:52  lr: 0.000549  loss: 2.8533 (2.9456)  time: 0.3458  data: 0.0002  max mem: 17867
[21:32:51.224219] Epoch: [107]  [5003/5004]  eta: 0:00:00  lr: 0.000547  loss: 2.7470 (2.9458)  time: 0.3457  data: 0.0011  max mem: 17867
[21:32:51.593569] Epoch: [107] Total time: 0:29:14 (0.3505 s / it)
[21:32:51.596752] Averaged stats: lr: 0.000547  loss: 2.7470 (2.9459)
[21:32:52.681189] Test:  [   0/1563]  eta: 0:28:09  loss: 0.3192 (0.3192)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.0810  data: 0.8932  max mem: 17867
[21:33:57.814746] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.8049 (0.9305)  acc1: 75.0000 (76.5469)  acc5: 93.7500 (94.7168)  time: 0.1300  data: 0.0003  max mem: 17867
[21:35:02.841894] Test:  [1000/1563]  eta: 0:01:13  loss: 1.2049 (1.0788)  acc1: 71.8750 (73.3423)  acc5: 90.6250 (92.5387)  time: 0.1300  data: 0.0002  max mem: 17867
[21:36:07.853253] Test:  [1500/1563]  eta: 0:00:08  loss: 0.6577 (1.1741)  acc1: 84.3750 (71.4149)  acc5: 96.8750 (91.1871)  time: 0.1299  data: 0.0002  max mem: 17867
[21:36:15.837610] Test:  [1562/1563]  eta: 0:00:00  loss: 0.5874 (1.1732)  acc1: 87.5000 (71.4580)  acc5: 96.8750 (91.2040)  time: 0.1262  data: 0.0001  max mem: 17867
[21:36:15.911904] Test: Total time: 0:03:24 (0.1307 s / it)
[21:36:16.056116] * Acc@1 71.458 Acc@5 91.204 loss 1.173
[21:36:16.056336] Accuracy of the network on the 50000 test images: 71.5%
[21:36:16.056376] Max accuracy: 71.74%
[21:36:16.235936] log_dir: ./output_dir_cml_spikformer
[21:36:17.793445] Epoch: [108]  [   0/5004]  eta: 2:09:49  lr: 0.000547  loss: 2.7247 (2.7247)  time: 1.5567  data: 1.2174  max mem: 17867
[21:47:57.670963] Epoch: [108]  [2000/5004]  eta: 0:17:32  lr: 0.000544  loss: 3.0298 (2.9471)  time: 0.3475  data: 0.0002  max mem: 17867
[21:59:37.092146] Epoch: [108]  [4000/5004]  eta: 0:05:51  lr: 0.000540  loss: 2.8062 (2.9463)  time: 0.3474  data: 0.0002  max mem: 17867
[22:05:28.239704] Epoch: [108]  [5003/5004]  eta: 0:00:00  lr: 0.000538  loss: 2.9552 (2.9435)  time: 0.3449  data: 0.0006  max mem: 17867
[22:05:28.602210] Epoch: [108] Total time: 0:29:12 (0.3502 s / it)
[22:05:28.604757] Averaged stats: lr: 0.000538  loss: 2.9552 (2.9358)
[22:05:29.736337] Test:  [   0/1563]  eta: 0:29:23  loss: 0.4790 (0.4790)  acc1: 90.6250 (90.6250)  acc5: 96.8750 (96.8750)  time: 1.1281  data: 0.9917  max mem: 17867
[22:06:34.792569] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.9423 (0.9405)  acc1: 75.0000 (75.8857)  acc5: 96.8750 (94.6357)  time: 0.1306  data: 0.0002  max mem: 17867
[22:07:39.758140] Test:  [1000/1563]  eta: 0:01:13  loss: 1.3361 (1.0896)  acc1: 62.5000 (72.9083)  acc5: 90.6250 (92.3295)  time: 0.1300  data: 0.0002  max mem: 17867
[22:08:44.743509] Test:  [1500/1563]  eta: 0:00:08  loss: 0.7448 (1.1806)  acc1: 81.2500 (70.9943)  acc5: 93.7500 (90.9935)  time: 0.1300  data: 0.0002  max mem: 17867
[22:08:52.722708] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4404 (1.1775)  acc1: 87.5000 (71.0700)  acc5: 96.8750 (91.0460)  time: 0.1262  data: 0.0001  max mem: 17867
[22:08:52.803234] Test: Total time: 0:03:24 (0.1306 s / it)
[22:08:52.919629] * Acc@1 71.070 Acc@5 91.046 loss 1.178
[22:08:52.919778] Accuracy of the network on the 50000 test images: 71.1%
[22:08:52.919799] Max accuracy: 71.74%
[22:08:52.943606] log_dir: ./output_dir_cml_spikformer
[22:08:54.473715] Epoch: [109]  [   0/5004]  eta: 2:07:31  lr: 0.000538  loss: 2.7538 (2.7538)  time: 1.5290  data: 1.1106  max mem: 17867
[22:20:34.423291] Epoch: [109]  [2000/5004]  eta: 0:17:33  lr: 0.000534  loss: 2.8978 (2.9205)  time: 0.3514  data: 0.0002  max mem: 17867
[22:32:13.605195] Epoch: [109]  [4000/5004]  eta: 0:05:51  lr: 0.000530  loss: 2.8821 (2.9269)  time: 0.3511  data: 0.0002  max mem: 17867
[22:38:04.653840] Epoch: [109]  [5003/5004]  eta: 0:00:00  lr: 0.000528  loss: 2.8233 (2.9262)  time: 0.3511  data: 0.0012  max mem: 17867
[22:38:04.966891] Epoch: [109] Total time: 0:29:12 (0.3501 s / it)
[22:38:04.993378] Averaged stats: lr: 0.000528  loss: 2.8233 (2.9323)
[22:38:06.228966] Test:  [   0/1563]  eta: 0:32:05  loss: 0.5508 (0.5508)  acc1: 84.3750 (84.3750)  acc5: 96.8750 (96.8750)  time: 1.2319  data: 1.0965  max mem: 17867
[22:39:11.245925] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.8270 (0.9111)  acc1: 75.0000 (76.8338)  acc5: 96.8750 (95.1160)  time: 0.1301  data: 0.0002  max mem: 17867
[22:40:16.257317] Test:  [1000/1563]  eta: 0:01:13  loss: 1.0894 (1.0434)  acc1: 68.7500 (74.4225)  acc5: 90.6250 (93.0320)  time: 0.1299  data: 0.0002  max mem: 17867
[22:41:21.256922] Test:  [1500/1563]  eta: 0:00:08  loss: 0.6132 (1.1319)  acc1: 84.3750 (72.4267)  acc5: 93.7500 (91.7472)  time: 0.1301  data: 0.0002  max mem: 17867
[22:41:29.251691] Test:  [1562/1563]  eta: 0:00:00  loss: 0.6637 (1.1335)  acc1: 87.5000 (72.3740)  acc5: 100.0000 (91.7380)  time: 0.1262  data: 0.0001  max mem: 17867
[22:41:29.317804] Test: Total time: 0:03:24 (0.1307 s / it)
[22:41:29.611864] * Acc@1 72.374 Acc@5 91.738 loss 1.133
[22:41:29.612067] Accuracy of the network on the 50000 test images: 72.4%
[22:41:29.612090] Max accuracy: 72.37%
[22:41:29.762034] log_dir: ./output_dir_cml_spikformer
[22:41:31.509431] Epoch: [110]  [   0/5004]  eta: 2:25:40  lr: 0.000528  loss: 3.0246 (3.0246)  time: 1.7466  data: 1.4183  max mem: 17867
[22:53:11.716776] Epoch: [110]  [2000/5004]  eta: 0:17:33  lr: 0.000524  loss: 2.7623 (2.9198)  time: 0.3523  data: 0.0002  max mem: 17867
[23:04:51.331889] Epoch: [110]  [4000/5004]  eta: 0:05:51  lr: 0.000521  loss: 2.8915 (2.9214)  time: 0.3473  data: 0.0002  max mem: 17867
[23:10:41.437681] Epoch: [110]  [5003/5004]  eta: 0:00:00  lr: 0.000519  loss: 2.9962 (2.9271)  time: 0.3461  data: 0.0011  max mem: 17867
[23:10:41.792905] Epoch: [110] Total time: 0:29:12 (0.3501 s / it)
[23:10:41.800860] Averaged stats: lr: 0.000519  loss: 2.9962 (2.9243)
[23:10:42.875212] Test:  [   0/1563]  eta: 0:27:53  loss: 0.4189 (0.4189)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.0707  data: 0.9327  max mem: 17867
[23:11:47.890691] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.8849 (0.9392)  acc1: 71.8750 (76.5719)  acc5: 93.7500 (94.8540)  time: 0.1299  data: 0.0002  max mem: 17867
[23:12:52.874089] Test:  [1000/1563]  eta: 0:01:13  loss: 1.4521 (1.0680)  acc1: 59.3750 (73.5171)  acc5: 87.5000 (92.7978)  time: 0.1299  data: 0.0002  max mem: 17867
[23:13:57.846337] Test:  [1500/1563]  eta: 0:00:08  loss: 0.6765 (1.1487)  acc1: 84.3750 (71.7501)  acc5: 96.8750 (91.5452)  time: 0.1299  data: 0.0002  max mem: 17867
[23:14:05.834793] Test:  [1562/1563]  eta: 0:00:00  loss: 0.3675 (1.1501)  acc1: 90.6250 (71.7160)  acc5: 96.8750 (91.5440)  time: 0.1262  data: 0.0001  max mem: 17867
[23:14:05.905482] Test: Total time: 0:03:24 (0.1306 s / it)
[23:14:06.015404] * Acc@1 71.716 Acc@5 91.544 loss 1.150
[23:14:06.015562] Accuracy of the network on the 50000 test images: 71.7%
[23:14:06.015583] Max accuracy: 72.37%
[23:14:06.049853] log_dir: ./output_dir_cml_spikformer
[23:14:07.528433] Epoch: [111]  [   0/5004]  eta: 2:03:11  lr: 0.000519  loss: 3.1576 (3.1576)  time: 1.4771  data: 1.1445  max mem: 17867
[23:25:47.157173] Epoch: [111]  [2000/5004]  eta: 0:17:32  lr: 0.000515  loss: 3.0230 (2.9124)  time: 0.3483  data: 0.0003  max mem: 17867
[23:37:26.479517] Epoch: [111]  [4000/5004]  eta: 0:05:51  lr: 0.000511  loss: 3.0824 (2.9193)  time: 0.3503  data: 0.0002  max mem: 17867
[23:43:17.213752] Epoch: [111]  [5003/5004]  eta: 0:00:00  lr: 0.000509  loss: 2.8373 (2.9179)  time: 0.3434  data: 0.0011  max mem: 17867
[23:43:17.570386] Epoch: [111] Total time: 0:29:11 (0.3500 s / it)
[23:43:17.573671] Averaged stats: lr: 0.000509  loss: 2.8373 (2.9165)
[23:43:19.022981] Test:  [   0/1563]  eta: 0:37:37  loss: 0.4582 (0.4582)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.4446  data: 1.2885  max mem: 17867
[23:44:24.074698] Test:  [ 500/1563]  eta: 0:02:21  loss: 1.0047 (0.9094)  acc1: 71.8750 (76.8900)  acc5: 93.7500 (95.0536)  time: 0.1299  data: 0.0002  max mem: 17867
[23:45:29.063981] Test:  [1000/1563]  eta: 0:01:13  loss: 1.2538 (1.0521)  acc1: 65.6250 (73.9635)  acc5: 93.7500 (92.9352)  time: 0.1299  data: 0.0002  max mem: 17867
[23:46:34.068409] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5816 (1.1404)  acc1: 84.3750 (71.9104)  acc5: 93.7500 (91.7222)  time: 0.1299  data: 0.0002  max mem: 17867
[23:46:42.060032] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4345 (1.1413)  acc1: 90.6250 (71.9180)  acc5: 96.8750 (91.6920)  time: 0.1263  data: 0.0001  max mem: 17867
[23:46:42.144747] Test: Total time: 0:03:24 (0.1309 s / it)
[23:46:42.236383] * Acc@1 71.918 Acc@5 91.692 loss 1.141
[23:46:42.236551] Accuracy of the network on the 50000 test images: 71.9%
[23:46:42.236571] Max accuracy: 72.37%
[23:46:42.244824] log_dir: ./output_dir_cml_spikformer
[23:46:43.742564] Epoch: [112]  [   0/5004]  eta: 2:04:50  lr: 0.000509  loss: 2.8285 (2.8285)  time: 1.4970  data: 1.0784  max mem: 17867
[23:58:25.043077] Epoch: [112]  [2000/5004]  eta: 0:17:35  lr: 0.000505  loss: 2.9522 (2.9041)  time: 0.3571  data: 0.0002  max mem: 17867
[00:10:05.218485] Epoch: [112]  [4000/5004]  eta: 0:05:52  lr: 0.000501  loss: 2.7938 (2.9062)  time: 0.3502  data: 0.0002  max mem: 17867
[00:15:56.303135] Epoch: [112]  [5003/5004]  eta: 0:00:00  lr: 0.000500  loss: 2.8862 (2.9084)  time: 0.3441  data: 0.0011  max mem: 17867
[00:15:56.779177] Epoch: [112] Total time: 0:29:14 (0.3506 s / it)
[00:15:56.779945] Averaged stats: lr: 0.000500  loss: 2.8862 (2.9096)
[00:15:58.177570] Test:  [   0/1563]  eta: 0:36:18  loss: 0.4997 (0.4997)  acc1: 90.6250 (90.6250)  acc5: 96.8750 (96.8750)  time: 1.3941  data: 1.2514  max mem: 17867
[00:17:03.205653] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.9225 (0.8854)  acc1: 75.0000 (77.5012)  acc5: 96.8750 (95.1035)  time: 0.1299  data: 0.0002  max mem: 17867
[00:18:08.199263] Test:  [1000/1563]  eta: 0:01:13  loss: 1.2045 (1.0437)  acc1: 68.7500 (74.2289)  acc5: 90.6250 (92.8540)  time: 0.1303  data: 0.0002  max mem: 17867
[00:19:13.198974] Test:  [1500/1563]  eta: 0:00:08  loss: 0.4985 (1.1279)  acc1: 87.5000 (72.4059)  acc5: 96.8750 (91.5910)  time: 0.1299  data: 0.0002  max mem: 17867
[00:19:21.181593] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4776 (1.1273)  acc1: 87.5000 (72.3760)  acc5: 100.0000 (91.6240)  time: 0.1262  data: 0.0001  max mem: 17867
[00:19:21.246011] Test: Total time: 0:03:24 (0.1308 s / it)
[00:19:21.440774] * Acc@1 72.376 Acc@5 91.624 loss 1.127
[00:19:21.440932] Accuracy of the network on the 50000 test images: 72.4%
[00:19:21.440954] Max accuracy: 72.38%
[00:19:21.448696] log_dir: ./output_dir_cml_spikformer
[00:19:22.928103] Epoch: [113]  [   0/5004]  eta: 2:03:17  lr: 0.000500  loss: 3.0346 (3.0346)  time: 1.4782  data: 1.1407  max mem: 17867
[00:31:03.660684] Epoch: [113]  [2000/5004]  eta: 0:17:34  lr: 0.000496  loss: 2.7518 (2.9000)  time: 0.3493  data: 0.0002  max mem: 17867
[00:42:43.306634] Epoch: [113]  [4000/5004]  eta: 0:05:51  lr: 0.000492  loss: 3.0048 (2.9049)  time: 0.3474  data: 0.0002  max mem: 17867
[00:48:34.331806] Epoch: [113]  [5003/5004]  eta: 0:00:00  lr: 0.000490  loss: 2.8258 (2.9054)  time: 0.3457  data: 0.0011  max mem: 17867
[00:48:34.694356] Epoch: [113] Total time: 0:29:13 (0.3504 s / it)
[00:48:34.702733] Averaged stats: lr: 0.000490  loss: 2.8258 (2.9060)
[00:48:36.035266] Test:  [   0/1563]  eta: 0:34:35  loss: 0.3170 (0.3170)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.3279  data: 0.8956  max mem: 17867
[00:49:41.248467] Test:  [ 500/1563]  eta: 0:02:21  loss: 0.9944 (0.9125)  acc1: 71.8750 (76.3785)  acc5: 93.7500 (94.7293)  time: 0.1300  data: 0.0002  max mem: 17867
[00:50:46.471829] Test:  [1000/1563]  eta: 0:01:14  loss: 1.5197 (1.0530)  acc1: 59.3750 (73.6264)  acc5: 87.5000 (92.5793)  time: 0.1302  data: 0.0002  max mem: 17867
[00:51:51.429040] Test:  [1500/1563]  eta: 0:00:08  loss: 0.6177 (1.1473)  acc1: 84.3750 (71.6668)  acc5: 96.8750 (91.3349)  time: 0.1300  data: 0.0002  max mem: 17867
[00:51:59.415462] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4994 (1.1448)  acc1: 90.6250 (71.7200)  acc5: 96.8750 (91.3900)  time: 0.1262  data: 0.0001  max mem: 17867
[00:51:59.481004] Test: Total time: 0:03:24 (0.1310 s / it)
[00:51:59.482241] * Acc@1 71.720 Acc@5 91.390 loss 1.145
[00:51:59.482355] Accuracy of the network on the 50000 test images: 71.7%
[00:51:59.482376] Max accuracy: 72.38%
[00:51:59.506748] log_dir: ./output_dir_cml_spikformer
[00:52:01.001897] Epoch: [114]  [   0/5004]  eta: 2:04:34  lr: 0.000490  loss: 3.0137 (3.0137)  time: 1.4936  data: 1.1604  max mem: 17867
[01:03:41.331107] Epoch: [114]  [2000/5004]  eta: 0:17:33  lr: 0.000486  loss: 2.7665 (2.8918)  time: 0.3452  data: 0.0002  max mem: 17867
[01:15:21.924644] Epoch: [114]  [4000/5004]  eta: 0:05:51  lr: 0.000482  loss: 2.8582 (2.8968)  time: 0.3494  data: 0.0002  max mem: 17867
[01:21:13.081097] Epoch: [114]  [5003/5004]  eta: 0:00:00  lr: 0.000481  loss: 2.8844 (2.8970)  time: 0.3452  data: 0.0011  max mem: 17867
[01:21:13.477808] Epoch: [114] Total time: 0:29:13 (0.3505 s / it)
[01:21:13.484042] Averaged stats: lr: 0.000481  loss: 2.8844 (2.8972)
[01:21:14.555093] Test:  [   0/1563]  eta: 0:27:48  loss: 0.4966 (0.4966)  acc1: 90.6250 (90.6250)  acc5: 96.8750 (96.8750)  time: 1.0676  data: 0.9104  max mem: 17867
[01:22:19.632904] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.9561 (0.8746)  acc1: 71.8750 (77.0210)  acc5: 96.8750 (95.1410)  time: 0.1314  data: 0.0002  max mem: 17867
[01:23:24.741479] Test:  [1000/1563]  eta: 0:01:13  loss: 1.3422 (1.0292)  acc1: 65.6250 (74.0104)  acc5: 90.6250 (92.9227)  time: 0.1300  data: 0.0002  max mem: 17867
[01:24:29.651256] Test:  [1500/1563]  eta: 0:00:08  loss: 0.6701 (1.1185)  acc1: 81.2500 (72.1894)  acc5: 93.7500 (91.6930)  time: 0.1298  data: 0.0002  max mem: 17867
[01:24:37.630938] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4729 (1.1194)  acc1: 90.6250 (72.2060)  acc5: 96.8750 (91.7140)  time: 0.1262  data: 0.0001  max mem: 17867
[01:24:37.692374] Test: Total time: 0:03:24 (0.1306 s / it)
[01:24:37.859184] * Acc@1 72.206 Acc@5 91.714 loss 1.119
[01:24:37.859323] Accuracy of the network on the 50000 test images: 72.2%
[01:24:37.859344] Max accuracy: 72.38%
[01:24:37.882739] log_dir: ./output_dir_cml_spikformer
[01:24:39.311502] Epoch: [115]  [   0/5004]  eta: 1:59:05  lr: 0.000481  loss: 3.2496 (3.2496)  time: 1.4279  data: 1.0383  max mem: 17867
[01:36:20.139068] Epoch: [115]  [2000/5004]  eta: 0:17:34  lr: 0.000477  loss: 2.9507 (2.8763)  time: 0.3565  data: 0.0002  max mem: 17867
[01:48:01.022851] Epoch: [115]  [4000/5004]  eta: 0:05:52  lr: 0.000473  loss: 2.8440 (2.8822)  time: 0.3479  data: 0.0002  max mem: 17867
[01:53:52.492311] Epoch: [115]  [5003/5004]  eta: 0:00:00  lr: 0.000471  loss: 2.8955 (2.8857)  time: 0.3479  data: 0.0011  max mem: 17867
[01:53:52.844348] Epoch: [115] Total time: 0:29:14 (0.3507 s / it)
[01:53:52.847436] Averaged stats: lr: 0.000471  loss: 2.8955 (2.8905)
[01:53:53.996988] Test:  [   0/1563]  eta: 0:29:51  loss: 0.5058 (0.5058)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.1460  data: 0.9731  max mem: 17867
[01:54:59.078333] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.9470 (0.9247)  acc1: 75.0000 (77.2954)  acc5: 96.8750 (94.9663)  time: 0.1299  data: 0.0002  max mem: 17867
[01:56:04.172771] Test:  [1000/1563]  eta: 0:01:13  loss: 1.5905 (1.0620)  acc1: 56.2500 (74.1290)  acc5: 90.6250 (92.9102)  time: 0.1308  data: 0.0002  max mem: 17867
[01:57:09.237592] Test:  [1500/1563]  eta: 0:00:08  loss: 0.6496 (1.1501)  acc1: 84.3750 (72.2039)  acc5: 96.8750 (91.6514)  time: 0.1299  data: 0.0002  max mem: 17867
[01:57:17.225181] Test:  [1562/1563]  eta: 0:00:00  loss: 0.5423 (1.1457)  acc1: 90.6250 (72.2980)  acc5: 96.8750 (91.6960)  time: 0.1263  data: 0.0001  max mem: 17867
[01:57:17.319296] Test: Total time: 0:03:24 (0.1308 s / it)
[01:57:17.347631] * Acc@1 72.298 Acc@5 91.696 loss 1.146
[01:57:17.347767] Accuracy of the network on the 50000 test images: 72.3%
[01:57:17.347789] Max accuracy: 72.38%
[01:57:17.405240] log_dir: ./output_dir_cml_spikformer
[01:57:18.889864] Epoch: [116]  [   0/5004]  eta: 2:03:45  lr: 0.000471  loss: 3.3735 (3.3735)  time: 1.4839  data: 1.0195  max mem: 17867
[02:09:00.040080] Epoch: [116]  [2000/5004]  eta: 0:17:34  lr: 0.000467  loss: 2.8921 (2.8849)  time: 0.3546  data: 0.0002  max mem: 17867
[02:20:41.043493] Epoch: [116]  [4000/5004]  eta: 0:05:52  lr: 0.000464  loss: 2.8336 (2.8848)  time: 0.3471  data: 0.0002  max mem: 17867
[02:26:32.125946] Epoch: [116]  [5003/5004]  eta: 0:00:00  lr: 0.000462  loss: 2.8103 (2.8853)  time: 0.3442  data: 0.0011  max mem: 17867
[02:26:32.494487] Epoch: [116] Total time: 0:29:15 (0.3507 s / it)
[02:26:32.499064] Averaged stats: lr: 0.000462  loss: 2.8103 (2.8870)
[02:26:33.493651] Test:  [   0/1563]  eta: 0:25:48  loss: 0.5509 (0.5509)  acc1: 90.6250 (90.6250)  acc5: 96.8750 (96.8750)  time: 0.9910  data: 0.8399  max mem: 17867
[02:27:38.511624] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.9259 (0.8853)  acc1: 75.0000 (77.9254)  acc5: 96.8750 (95.3094)  time: 0.1300  data: 0.0002  max mem: 17867
[02:28:43.512865] Test:  [1000/1563]  eta: 0:01:13  loss: 1.2665 (1.0357)  acc1: 71.8750 (74.7003)  acc5: 93.7500 (93.0694)  time: 0.1299  data: 0.0002  max mem: 17867
[02:29:48.508724] Test:  [1500/1563]  eta: 0:00:08  loss: 0.6933 (1.1176)  acc1: 81.2500 (72.9701)  acc5: 96.8750 (91.9637)  time: 0.1299  data: 0.0002  max mem: 17867
[02:29:56.538405] Test:  [1562/1563]  eta: 0:00:00  loss: 0.5061 (1.1194)  acc1: 87.5000 (72.9140)  acc5: 96.8750 (91.9580)  time: 0.1285  data: 0.0001  max mem: 17867
[02:29:56.610692] Test: Total time: 0:03:24 (0.1306 s / it)
[02:29:56.779579] * Acc@1 72.914 Acc@5 91.958 loss 1.119
[02:29:56.779715] Accuracy of the network on the 50000 test images: 72.9%
[02:29:56.779745] Max accuracy: 72.91%
[02:29:56.787458] log_dir: ./output_dir_cml_spikformer
[02:29:58.289758] Epoch: [117]  [   0/5004]  eta: 2:05:13  lr: 0.000462  loss: 2.5724 (2.5724)  time: 1.5015  data: 1.1075  max mem: 17867
[02:41:39.581802] Epoch: [117]  [2000/5004]  eta: 0:17:35  lr: 0.000458  loss: 2.9788 (2.8696)  time: 0.3600  data: 0.0002  max mem: 17867
[02:53:20.488209] Epoch: [117]  [4000/5004]  eta: 0:05:52  lr: 0.000454  loss: 2.9026 (2.8719)  time: 0.3508  data: 0.0002  max mem: 17867
[02:59:11.244793] Epoch: [117]  [5003/5004]  eta: 0:00:00  lr: 0.000452  loss: 2.8035 (2.8744)  time: 0.3439  data: 0.0013  max mem: 17867
[02:59:11.596905] Epoch: [117] Total time: 0:29:14 (0.3507 s / it)
[02:59:11.607174] Averaged stats: lr: 0.000452  loss: 2.8035 (2.8787)
[02:59:12.628785] Test:  [   0/1563]  eta: 0:26:30  loss: 0.2832 (0.2832)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.0177  data: 0.8642  max mem: 17867
[03:00:17.586717] Test:  [ 500/1563]  eta: 0:02:19  loss: 0.8959 (0.8897)  acc1: 75.0000 (77.8318)  acc5: 96.8750 (95.4528)  time: 0.1299  data: 0.0002  max mem: 17867
[03:01:22.540674] Test:  [1000/1563]  eta: 0:01:13  loss: 0.9797 (1.0363)  acc1: 75.0000 (74.5973)  acc5: 90.6250 (93.2286)  time: 0.1298  data: 0.0002  max mem: 17867
[03:02:27.486154] Test:  [1500/1563]  eta: 0:00:08  loss: 0.6415 (1.1212)  acc1: 84.3750 (72.7744)  acc5: 96.8750 (91.9699)  time: 0.1299  data: 0.0002  max mem: 17867
[03:02:35.467746] Test:  [1562/1563]  eta: 0:00:00  loss: 0.5035 (1.1212)  acc1: 90.6250 (72.7800)  acc5: 96.8750 (91.9880)  time: 0.1261  data: 0.0001  max mem: 17867
[03:02:35.525836] Test: Total time: 0:03:23 (0.1305 s / it)
[03:02:35.899696] * Acc@1 72.780 Acc@5 91.988 loss 1.121
[03:02:35.899843] Accuracy of the network on the 50000 test images: 72.8%
[03:02:35.899864] Max accuracy: 72.91%
[03:02:35.906725] log_dir: ./output_dir_cml_spikformer
[03:02:37.431184] Epoch: [118]  [   0/5004]  eta: 2:07:01  lr: 0.000452  loss: 3.2344 (3.2344)  time: 1.5231  data: 0.9939  max mem: 17867
[03:14:18.209301] Epoch: [118]  [2000/5004]  eta: 0:17:34  lr: 0.000449  loss: 2.9831 (2.8629)  time: 0.3508  data: 0.0002  max mem: 17867
[03:25:58.019925] Epoch: [118]  [4000/5004]  eta: 0:05:51  lr: 0.000445  loss: 2.8359 (2.8727)  time: 0.3489  data: 0.0002  max mem: 17867
[03:31:48.979633] Epoch: [118]  [5003/5004]  eta: 0:00:00  lr: 0.000443  loss: 2.7872 (2.8762)  time: 0.3446  data: 0.0011  max mem: 17867
[03:31:49.355296] Epoch: [118] Total time: 0:29:13 (0.3504 s / it)
[03:31:49.364868] Averaged stats: lr: 0.000443  loss: 2.7872 (2.8728)
[03:31:50.448965] Test:  [   0/1563]  eta: 0:28:08  loss: 0.3570 (0.3570)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.0805  data: 0.9179  max mem: 17867
[03:32:55.537764] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.8273 (0.9138)  acc1: 78.1250 (77.2455)  acc5: 96.8750 (94.7979)  time: 0.1299  data: 0.0002  max mem: 17867
[03:34:00.519140] Test:  [1000/1563]  eta: 0:01:13  loss: 1.4481 (1.0463)  acc1: 65.6250 (74.2539)  acc5: 87.5000 (92.8009)  time: 0.1300  data: 0.0002  max mem: 17867
[03:35:05.523711] Test:  [1500/1563]  eta: 0:00:08  loss: 0.6574 (1.1255)  acc1: 87.5000 (72.4496)  acc5: 96.8750 (91.7284)  time: 0.1299  data: 0.0002  max mem: 17867
[03:35:13.508187] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4948 (1.1259)  acc1: 90.6250 (72.4120)  acc5: 96.8750 (91.7420)  time: 0.1262  data: 0.0001  max mem: 17867
[03:35:13.585346] Test: Total time: 0:03:24 (0.1307 s / it)
[03:35:13.713807] * Acc@1 72.412 Acc@5 91.742 loss 1.126
[03:35:13.713955] Accuracy of the network on the 50000 test images: 72.4%
[03:35:13.713982] Max accuracy: 72.91%
[03:35:13.769823] log_dir: ./output_dir_cml_spikformer
[03:35:15.355459] Epoch: [119]  [   0/5004]  eta: 2:12:07  lr: 0.000443  loss: 2.9939 (2.9939)  time: 1.5843  data: 1.0964  max mem: 17867
[03:46:55.868978] Epoch: [119]  [2000/5004]  eta: 0:17:33  lr: 0.000439  loss: 2.7061 (2.8522)  time: 0.3451  data: 0.0002  max mem: 17867
[03:58:35.710808] Epoch: [119]  [4000/5004]  eta: 0:05:51  lr: 0.000436  loss: 2.7924 (2.8599)  time: 0.3535  data: 0.0002  max mem: 17867
[04:04:26.587501] Epoch: [119]  [5003/5004]  eta: 0:00:00  lr: 0.000434  loss: 2.7732 (2.8636)  time: 0.3428  data: 0.0006  max mem: 17867
[04:04:26.959007] Epoch: [119] Total time: 0:29:13 (0.3504 s / it)
[04:04:26.960110] Averaged stats: lr: 0.000434  loss: 2.7732 (2.8654)
[04:04:28.036316] Test:  [   0/1563]  eta: 0:27:56  loss: 0.3216 (0.3216)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.0726  data: 0.9334  max mem: 17867
[04:05:33.055046] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.7451 (0.8704)  acc1: 81.2500 (77.4014)  acc5: 96.8750 (95.4591)  time: 0.1300  data: 0.0002  max mem: 17867
[04:06:38.057409] Test:  [1000/1563]  eta: 0:01:13  loss: 1.3246 (1.0280)  acc1: 68.7500 (74.4225)  acc5: 93.7500 (93.0757)  time: 0.1300  data: 0.0002  max mem: 17867
[04:07:43.039549] Test:  [1500/1563]  eta: 0:00:08  loss: 0.6248 (1.1070)  acc1: 84.3750 (72.7848)  acc5: 96.8750 (91.9033)  time: 0.1299  data: 0.0002  max mem: 17867
[04:07:51.021937] Test:  [1562/1563]  eta: 0:00:00  loss: 0.5114 (1.1046)  acc1: 87.5000 (72.8280)  acc5: 96.8750 (91.9620)  time: 0.1262  data: 0.0001  max mem: 17867
[04:07:51.100912] Test: Total time: 0:03:24 (0.1306 s / it)
[04:07:51.284471] * Acc@1 72.828 Acc@5 91.962 loss 1.105
[04:07:51.284612] Accuracy of the network on the 50000 test images: 72.8%
[04:07:51.284632] Max accuracy: 72.91%
[04:07:51.310480] log_dir: ./output_dir_cml_spikformer
[04:07:52.902488] Epoch: [120]  [   0/5004]  eta: 2:12:43  lr: 0.000434  loss: 3.0199 (3.0199)  time: 1.5914  data: 1.1772  max mem: 17867
[04:19:32.554765] Epoch: [120]  [2000/5004]  eta: 0:17:32  lr: 0.000430  loss: 2.8131 (2.8518)  time: 0.3507  data: 0.0002  max mem: 17867
[04:31:11.017726] Epoch: [120]  [4000/5004]  eta: 0:05:51  lr: 0.000426  loss: 2.8850 (2.8550)  time: 0.3491  data: 0.0002  max mem: 17867
[04:37:02.014977] Epoch: [120]  [5003/5004]  eta: 0:00:00  lr: 0.000424  loss: 2.6423 (2.8602)  time: 0.3470  data: 0.0007  max mem: 17867
[04:37:02.390855] Epoch: [120] Total time: 0:29:11 (0.3499 s / it)
[04:37:02.391752] Averaged stats: lr: 0.000424  loss: 2.6423 (2.8575)
[04:37:03.529510] Test:  [   0/1563]  eta: 0:29:28  loss: 0.3743 (0.3743)  acc1: 90.6250 (90.6250)  acc5: 96.8750 (96.8750)  time: 1.1314  data: 0.9911  max mem: 17867
[04:38:08.517524] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.8200 (0.8669)  acc1: 78.1250 (77.9878)  acc5: 96.8750 (95.2408)  time: 0.1303  data: 0.0002  max mem: 17867
[04:39:13.479902] Test:  [1000/1563]  eta: 0:01:13  loss: 1.3887 (1.0186)  acc1: 68.7500 (74.7690)  acc5: 90.6250 (93.0569)  time: 0.1298  data: 0.0002  max mem: 17867
[04:40:18.462881] Test:  [1500/1563]  eta: 0:00:08  loss: 0.4735 (1.1028)  acc1: 84.3750 (72.8452)  acc5: 96.8750 (91.9345)  time: 0.1299  data: 0.0002  max mem: 17867
[04:40:26.466308] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4853 (1.1025)  acc1: 87.5000 (72.8380)  acc5: 100.0000 (91.9540)  time: 0.1261  data: 0.0001  max mem: 17867
[04:40:26.538214] Test: Total time: 0:03:24 (0.1306 s / it)
[04:40:26.924528] * Acc@1 72.838 Acc@5 91.954 loss 1.103
[04:40:26.924666] Accuracy of the network on the 50000 test images: 72.8%
[04:40:26.924687] Max accuracy: 72.91%
[04:40:27.025307] log_dir: ./output_dir_cml_spikformer
[04:40:28.534120] Epoch: [121]  [   0/5004]  eta: 2:05:45  lr: 0.000424  loss: 2.2130 (2.2130)  time: 1.5079  data: 1.1377  max mem: 17867
[04:52:07.291697] Epoch: [121]  [2000/5004]  eta: 0:17:31  lr: 0.000421  loss: 2.6326 (2.8407)  time: 0.3491  data: 0.0002  max mem: 17867
[05:03:45.933448] Epoch: [121]  [4000/5004]  eta: 0:05:51  lr: 0.000417  loss: 2.8051 (2.8460)  time: 0.3487  data: 0.0002  max mem: 17867
[05:09:36.044413] Epoch: [121]  [5003/5004]  eta: 0:00:00  lr: 0.000415  loss: 2.8621 (2.8483)  time: 0.3471  data: 0.0011  max mem: 17867
[05:09:36.406242] Epoch: [121] Total time: 0:29:09 (0.3496 s / it)
[05:09:36.415876] Averaged stats: lr: 0.000415  loss: 2.8621 (2.8514)
[05:09:37.431227] Test:  [   0/1563]  eta: 0:26:21  loss: 0.3610 (0.3610)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.0118  data: 0.8511  max mem: 17867
[05:10:42.454372] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.7885 (0.8913)  acc1: 78.1250 (77.4389)  acc5: 96.8750 (95.1035)  time: 0.1299  data: 0.0002  max mem: 17867
[05:11:47.447252] Test:  [1000/1563]  eta: 0:01:13  loss: 1.4826 (1.0312)  acc1: 59.3750 (74.6129)  acc5: 90.6250 (93.0538)  time: 0.1300  data: 0.0002  max mem: 17867
[05:12:52.449455] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5313 (1.1091)  acc1: 84.3750 (72.9118)  acc5: 96.8750 (92.0511)  time: 0.1299  data: 0.0002  max mem: 17867
[05:13:00.462465] Test:  [1562/1563]  eta: 0:00:00  loss: 0.5693 (1.1097)  acc1: 87.5000 (72.9080)  acc5: 96.8750 (92.0800)  time: 0.1262  data: 0.0001  max mem: 17867
[05:13:00.574440] Test: Total time: 0:03:24 (0.1306 s / it)
[05:13:00.672034] * Acc@1 72.908 Acc@5 92.080 loss 1.110
[05:13:00.672185] Accuracy of the network on the 50000 test images: 72.9%
[05:13:00.672209] Max accuracy: 72.91%
[05:13:00.697869] log_dir: ./output_dir_cml_spikformer
[05:13:02.197026] Epoch: [122]  [   0/5004]  eta: 2:04:56  lr: 0.000415  loss: 2.6361 (2.6361)  time: 1.4982  data: 1.1237  max mem: 17867
[05:24:41.309365] Epoch: [122]  [2000/5004]  eta: 0:17:31  lr: 0.000412  loss: 2.8471 (2.8342)  time: 0.3526  data: 0.0002  max mem: 17867
[05:36:20.328851] Epoch: [122]  [4000/5004]  eta: 0:05:51  lr: 0.000408  loss: 2.9485 (2.8408)  time: 0.3460  data: 0.0002  max mem: 17867
[05:42:10.467316] Epoch: [122]  [5003/5004]  eta: 0:00:00  lr: 0.000406  loss: 2.7364 (2.8405)  time: 0.3460  data: 0.0012  max mem: 17867
[05:42:10.811215] Epoch: [122] Total time: 0:29:10 (0.3497 s / it)
[05:42:10.819527] Averaged stats: lr: 0.000406  loss: 2.7364 (2.8455)
[05:42:12.150545] Test:  [   0/1563]  eta: 0:34:34  loss: 0.2989 (0.2989)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.3274  data: 1.1888  max mem: 17867
[05:43:17.198080] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.7918 (0.8680)  acc1: 78.1250 (78.0626)  acc5: 96.8750 (95.1909)  time: 0.1299  data: 0.0002  max mem: 17867
[05:44:22.246724] Test:  [1000/1563]  eta: 0:01:13  loss: 1.2454 (1.0139)  acc1: 68.7500 (74.5942)  acc5: 93.7500 (93.4253)  time: 0.1299  data: 0.0002  max mem: 17867
[05:45:27.252838] Test:  [1500/1563]  eta: 0:00:08  loss: 0.7186 (1.1052)  acc1: 84.3750 (72.7973)  acc5: 93.7500 (92.1073)  time: 0.1300  data: 0.0002  max mem: 17867
[05:45:35.236846] Test:  [1562/1563]  eta: 0:00:00  loss: 0.3669 (1.1037)  acc1: 90.6250 (72.8060)  acc5: 100.0000 (92.1340)  time: 0.1262  data: 0.0001  max mem: 17867
[05:45:35.331112] Test: Total time: 0:03:24 (0.1308 s / it)
[05:45:35.371277] * Acc@1 72.806 Acc@5 92.134 loss 1.104
[05:45:35.371475] Accuracy of the network on the 50000 test images: 72.8%
[05:45:35.371500] Max accuracy: 72.91%
[05:45:35.427681] log_dir: ./output_dir_cml_spikformer
[05:45:37.038935] Epoch: [123]  [   0/5004]  eta: 2:14:18  lr: 0.000406  loss: 2.5667 (2.5667)  time: 1.6104  data: 1.0473  max mem: 17867
[05:57:19.427373] Epoch: [123]  [2000/5004]  eta: 0:17:36  lr: 0.000402  loss: 2.7240 (2.8368)  time: 0.3508  data: 0.0002  max mem: 17867
[06:09:00.233640] Epoch: [123]  [4000/5004]  eta: 0:05:52  lr: 0.000399  loss: 3.0486 (2.8353)  time: 0.3534  data: 0.0002  max mem: 17867
[06:14:51.432542] Epoch: [123]  [5003/5004]  eta: 0:00:00  lr: 0.000397  loss: 2.9369 (2.8369)  time: 0.3450  data: 0.0011  max mem: 17867
[06:14:51.797418] Epoch: [123] Total time: 0:29:16 (0.3510 s / it)
[06:14:51.799616] Averaged stats: lr: 0.000397  loss: 2.9369 (2.8372)
[06:14:52.881355] Test:  [   0/1563]  eta: 0:28:02  loss: 0.3700 (0.3700)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.0767  data: 0.9214  max mem: 17867
[06:15:57.921061] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.0116 (0.8844)  acc1: 71.8750 (77.9441)  acc5: 96.8750 (95.1472)  time: 0.1299  data: 0.0002  max mem: 17867
[06:17:02.901651] Test:  [1000/1563]  eta: 0:01:13  loss: 1.2157 (1.0041)  acc1: 65.6250 (75.2217)  acc5: 93.7500 (93.5315)  time: 0.1299  data: 0.0002  max mem: 17867
[06:18:07.898224] Test:  [1500/1563]  eta: 0:00:08  loss: 0.6474 (1.0944)  acc1: 81.2500 (73.1658)  acc5: 93.7500 (92.2947)  time: 0.1299  data: 0.0002  max mem: 17867
[06:18:15.908068] Test:  [1562/1563]  eta: 0:00:00  loss: 0.5018 (1.0951)  acc1: 90.6250 (73.1700)  acc5: 96.8750 (92.2760)  time: 0.1275  data: 0.0001  max mem: 17867
[06:18:15.999808] Test: Total time: 0:03:24 (0.1306 s / it)
[06:18:16.114308] * Acc@1 73.170 Acc@5 92.276 loss 1.095
[06:18:16.114445] Accuracy of the network on the 50000 test images: 73.2%
[06:18:16.114468] Max accuracy: 73.17%
[06:18:16.125965] log_dir: ./output_dir_cml_spikformer
[06:18:17.716751] Epoch: [124]  [   0/5004]  eta: 2:12:35  lr: 0.000397  loss: 2.9854 (2.9854)  time: 1.5899  data: 1.0065  max mem: 17867
[06:29:59.007869] Epoch: [124]  [2000/5004]  eta: 0:17:35  lr: 0.000393  loss: 2.8586 (2.8331)  time: 0.3533  data: 0.0002  max mem: 17867
[06:41:39.344202] Epoch: [124]  [4000/5004]  eta: 0:05:52  lr: 0.000390  loss: 2.9650 (2.8294)  time: 0.3473  data: 0.0002  max mem: 17867
[06:47:29.824808] Epoch: [124]  [5003/5004]  eta: 0:00:00  lr: 0.000388  loss: 2.6842 (2.8297)  time: 0.3434  data: 0.0007  max mem: 17867
[06:47:30.202691] Epoch: [124] Total time: 0:29:14 (0.3505 s / it)
[06:47:30.203509] Averaged stats: lr: 0.000388  loss: 2.6842 (2.8335)
[06:47:31.286918] Test:  [   0/1563]  eta: 0:28:05  loss: 0.3975 (0.3975)  acc1: 93.7500 (93.7500)  acc5: 93.7500 (93.7500)  time: 1.0786  data: 0.9356  max mem: 17867
[06:48:36.283290] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.9266 (0.8374)  acc1: 78.1250 (78.5554)  acc5: 93.7500 (95.3842)  time: 0.1300  data: 0.0002  max mem: 17867
[06:49:41.257568] Test:  [1000/1563]  eta: 0:01:13  loss: 1.2572 (0.9847)  acc1: 65.6250 (75.2997)  acc5: 93.7500 (93.4409)  time: 0.1301  data: 0.0002  max mem: 17867
[06:50:46.223487] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5860 (1.0716)  acc1: 84.3750 (73.3199)  acc5: 96.8750 (92.3509)  time: 0.1298  data: 0.0002  max mem: 17867
[06:50:54.210617] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4252 (1.0726)  acc1: 90.6250 (73.2960)  acc5: 96.8750 (92.3280)  time: 0.1262  data: 0.0001  max mem: 17867
[06:50:54.272134] Test: Total time: 0:03:24 (0.1306 s / it)
[06:50:54.775134] * Acc@1 73.296 Acc@5 92.328 loss 1.073
[06:50:54.775270] Accuracy of the network on the 50000 test images: 73.3%
[06:50:54.775291] Max accuracy: 73.30%
[06:50:54.937261] log_dir: ./output_dir_cml_spikformer
[06:50:56.457627] Epoch: [125]  [   0/5004]  eta: 2:06:43  lr: 0.000388  loss: 3.0258 (3.0258)  time: 1.5195  data: 1.1394  max mem: 17867
[07:02:37.558470] Epoch: [125]  [2000/5004]  eta: 0:17:34  lr: 0.000384  loss: 2.7840 (2.8187)  time: 0.3491  data: 0.0002  max mem: 17867
[07:14:17.479468] Epoch: [125]  [4000/5004]  eta: 0:05:51  lr: 0.000381  loss: 2.9773 (2.8274)  time: 0.3533  data: 0.0002  max mem: 17867
[07:20:08.609728] Epoch: [125]  [5003/5004]  eta: 0:00:00  lr: 0.000379  loss: 2.8169 (2.8267)  time: 0.3483  data: 0.0006  max mem: 17867
[07:20:08.988672] Epoch: [125] Total time: 0:29:14 (0.3505 s / it)
[07:20:08.989790] Averaged stats: lr: 0.000379  loss: 2.8169 (2.8228)
[07:20:10.091867] Test:  [   0/1563]  eta: 0:28:33  loss: 0.6889 (0.6889)  acc1: 90.6250 (90.6250)  acc5: 93.7500 (93.7500)  time: 1.0961  data: 0.9208  max mem: 17867
[07:21:15.243216] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.1299 (0.8828)  acc1: 75.0000 (77.6135)  acc5: 90.6250 (95.2720)  time: 0.1299  data: 0.0002  max mem: 17867
[07:22:20.235096] Test:  [1000/1563]  eta: 0:01:13  loss: 1.3450 (1.0124)  acc1: 62.5000 (74.9157)  acc5: 90.6250 (93.4472)  time: 0.1299  data: 0.0002  max mem: 17867
[07:23:25.222745] Test:  [1500/1563]  eta: 0:00:08  loss: 0.6574 (1.0933)  acc1: 87.5000 (73.1679)  acc5: 96.8750 (92.3051)  time: 0.1301  data: 0.0002  max mem: 17867
[07:23:33.204779] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4803 (1.0959)  acc1: 90.6250 (73.1200)  acc5: 96.8750 (92.2760)  time: 0.1261  data: 0.0001  max mem: 17867
[07:23:33.282914] Test: Total time: 0:03:24 (0.1307 s / it)
[07:23:33.356757] * Acc@1 73.120 Acc@5 92.276 loss 1.096
[07:23:33.356891] Accuracy of the network on the 50000 test images: 73.1%
[07:23:33.356911] Max accuracy: 73.30%
[07:23:33.379413] log_dir: ./output_dir_cml_spikformer
[07:23:34.959723] Epoch: [126]  [   0/5004]  eta: 2:11:44  lr: 0.000379  loss: 2.4682 (2.4682)  time: 1.5797  data: 0.9847  max mem: 17867
[07:35:14.895050] Epoch: [126]  [2000/5004]  eta: 0:17:33  lr: 0.000375  loss: 2.7898 (2.8214)  time: 0.3494  data: 0.0002  max mem: 17867
[07:46:54.031718] Epoch: [126]  [4000/5004]  eta: 0:05:51  lr: 0.000372  loss: 2.7194 (2.8184)  time: 0.3495  data: 0.0002  max mem: 17867
[07:52:44.751287] Epoch: [126]  [5003/5004]  eta: 0:00:00  lr: 0.000370  loss: 2.7881 (2.8141)  time: 0.3441  data: 0.0012  max mem: 17867
[07:52:45.087949] Epoch: [126] Total time: 0:29:11 (0.3501 s / it)
[07:52:45.094030] Averaged stats: lr: 0.000370  loss: 2.7881 (2.8179)
[07:52:46.108651] Test:  [   0/1563]  eta: 0:26:20  loss: 0.3056 (0.3056)  acc1: 90.6250 (90.6250)  acc5: 96.8750 (96.8750)  time: 1.0111  data: 0.8673  max mem: 17867
[07:53:51.128375] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.8615 (0.8413)  acc1: 75.0000 (78.5679)  acc5: 93.7500 (95.3094)  time: 0.1298  data: 0.0002  max mem: 17867
[07:54:56.202099] Test:  [1000/1563]  eta: 0:01:13  loss: 1.1568 (0.9931)  acc1: 68.7500 (75.2685)  acc5: 90.6250 (93.3410)  time: 0.1312  data: 0.0002  max mem: 17867
[07:56:01.190155] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5222 (1.0769)  acc1: 81.2500 (73.3615)  acc5: 96.8750 (92.1927)  time: 0.1303  data: 0.0002  max mem: 17867
[07:56:09.201638] Test:  [1562/1563]  eta: 0:00:00  loss: 0.5501 (1.0757)  acc1: 90.6250 (73.3760)  acc5: 96.8750 (92.2040)  time: 0.1262  data: 0.0001  max mem: 17867
[07:56:09.272297] Test: Total time: 0:03:24 (0.1306 s / it)
[07:56:09.655606] * Acc@1 73.376 Acc@5 92.204 loss 1.076
[07:56:09.655747] Accuracy of the network on the 50000 test images: 73.4%
[07:56:09.655768] Max accuracy: 73.38%
[07:56:09.739020] log_dir: ./output_dir_cml_spikformer
[07:56:11.170208] Epoch: [127]  [   0/5004]  eta: 1:59:17  lr: 0.000370  loss: 2.6704 (2.6704)  time: 1.4304  data: 1.0393  max mem: 17867
[08:07:51.544932] Epoch: [127]  [2000/5004]  eta: 0:17:33  lr: 0.000366  loss: 2.8624 (2.8021)  time: 0.3537  data: 0.0002  max mem: 17867
[08:19:30.807711] Epoch: [127]  [4000/5004]  eta: 0:05:51  lr: 0.000363  loss: 2.8772 (2.8086)  time: 0.3474  data: 0.0002  max mem: 17867
[08:25:21.183135] Epoch: [127]  [5003/5004]  eta: 0:00:00  lr: 0.000361  loss: 2.6770 (2.8103)  time: 0.3473  data: 0.0011  max mem: 17867
[08:25:21.550895] Epoch: [127] Total time: 0:29:11 (0.3501 s / it)
[08:25:21.557230] Averaged stats: lr: 0.000361  loss: 2.6770 (2.8100)
[08:25:22.611929] Test:  [   0/1563]  eta: 0:27:22  loss: 0.4707 (0.4707)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.0510  data: 0.9121  max mem: 17867
[08:26:27.624318] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.8101 (0.8641)  acc1: 81.2500 (78.1437)  acc5: 96.8750 (95.3343)  time: 0.1303  data: 0.0002  max mem: 17867
[08:27:32.662585] Test:  [1000/1563]  eta: 0:01:13  loss: 1.3073 (0.9950)  acc1: 65.6250 (75.5089)  acc5: 90.6250 (93.4628)  time: 0.1299  data: 0.0002  max mem: 17867
[08:28:37.647374] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5422 (1.0820)  acc1: 87.5000 (73.6113)  acc5: 96.8750 (92.2281)  time: 0.1300  data: 0.0002  max mem: 17867
[08:28:45.632453] Test:  [1562/1563]  eta: 0:00:00  loss: 0.5578 (1.0838)  acc1: 87.5000 (73.5560)  acc5: 93.7500 (92.2060)  time: 0.1262  data: 0.0001  max mem: 17867
[08:28:45.726137] Test: Total time: 0:03:24 (0.1306 s / it)
[08:28:45.805341] * Acc@1 73.556 Acc@5 92.206 loss 1.084
[08:28:45.805479] Accuracy of the network on the 50000 test images: 73.6%
[08:28:45.805503] Max accuracy: 73.56%
[08:28:45.813735] log_dir: ./output_dir_cml_spikformer
[08:28:47.382079] Epoch: [128]  [   0/5004]  eta: 2:10:41  lr: 0.000361  loss: 2.2004 (2.2004)  time: 1.5671  data: 1.1062  max mem: 17867
[08:40:27.618663] Epoch: [128]  [2000/5004]  eta: 0:17:33  lr: 0.000358  loss: 2.6664 (2.7943)  time: 0.3469  data: 0.0002  max mem: 17867
[08:52:07.484153] Epoch: [128]  [4000/5004]  eta: 0:05:51  lr: 0.000354  loss: 2.8994 (2.7951)  time: 0.3579  data: 0.0002  max mem: 17867
[08:57:58.215975] Epoch: [128]  [5003/5004]  eta: 0:00:00  lr: 0.000352  loss: 2.9403 (2.8016)  time: 0.3448  data: 0.0006  max mem: 17867
[08:57:58.578577] Epoch: [128] Total time: 0:29:12 (0.3503 s / it)
[08:57:58.579359] Averaged stats: lr: 0.000352  loss: 2.9403 (2.8048)
[08:57:59.671611] Test:  [   0/1563]  eta: 0:28:20  loss: 0.3578 (0.3578)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.0878  data: 0.9107  max mem: 17867
[08:59:04.671869] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.8668 (0.8921)  acc1: 78.1250 (78.0501)  acc5: 96.8750 (95.1534)  time: 0.1299  data: 0.0002  max mem: 17867
[09:00:09.829758] Test:  [1000/1563]  eta: 0:01:13  loss: 1.2107 (1.0028)  acc1: 68.7500 (75.5932)  acc5: 93.7500 (93.4534)  time: 0.1312  data: 0.0002  max mem: 17867
[09:01:14.889634] Test:  [1500/1563]  eta: 0:00:08  loss: 0.6191 (1.0919)  acc1: 84.3750 (73.4427)  acc5: 96.8750 (92.3218)  time: 0.1299  data: 0.0002  max mem: 17867
[09:01:22.877511] Test:  [1562/1563]  eta: 0:00:00  loss: 0.3850 (1.0920)  acc1: 90.6250 (73.4400)  acc5: 96.8750 (92.3320)  time: 0.1263  data: 0.0001  max mem: 17867
[09:01:22.944112] Test: Total time: 0:03:24 (0.1307 s / it)
[09:01:22.945265] * Acc@1 73.440 Acc@5 92.332 loss 1.092
[09:01:22.945411] Accuracy of the network on the 50000 test images: 73.4%
[09:01:22.945437] Max accuracy: 73.56%
[09:01:22.981933] log_dir: ./output_dir_cml_spikformer
[09:01:24.627301] Epoch: [129]  [   0/5004]  eta: 2:17:10  lr: 0.000352  loss: 2.4551 (2.4551)  time: 1.6447  data: 1.1609  max mem: 17867
[09:13:04.096921] Epoch: [129]  [2000/5004]  eta: 0:17:32  lr: 0.000349  loss: 2.7078 (2.7963)  time: 0.3570  data: 0.0002  max mem: 17867
[09:24:43.456846] Epoch: [129]  [4000/5004]  eta: 0:05:51  lr: 0.000345  loss: 2.6573 (2.7972)  time: 0.3475  data: 0.0003  max mem: 17867
[09:30:34.081444] Epoch: [129]  [5003/5004]  eta: 0:00:00  lr: 0.000344  loss: 2.7984 (2.7996)  time: 0.3448  data: 0.0007  max mem: 17867
[09:30:34.442878] Epoch: [129] Total time: 0:29:11 (0.3500 s / it)
[09:30:34.450611] Averaged stats: lr: 0.000344  loss: 2.7984 (2.7934)
[09:30:35.546665] Test:  [   0/1563]  eta: 0:28:27  loss: 0.3107 (0.3107)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.0922  data: 0.9517  max mem: 17867
[09:31:40.569848] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.8252 (0.8130)  acc1: 75.0000 (79.0918)  acc5: 96.8750 (95.8271)  time: 0.1299  data: 0.0002  max mem: 17867
[09:32:45.597971] Test:  [1000/1563]  eta: 0:01:13  loss: 1.2804 (0.9779)  acc1: 59.3750 (75.6712)  acc5: 90.6250 (93.5283)  time: 0.1298  data: 0.0002  max mem: 17867
[09:33:50.609344] Test:  [1500/1563]  eta: 0:00:08  loss: 0.6042 (1.0562)  acc1: 87.5000 (74.1089)  acc5: 93.7500 (92.4904)  time: 0.1300  data: 0.0002  max mem: 17867
[09:33:58.592574] Test:  [1562/1563]  eta: 0:00:00  loss: 0.3671 (1.0584)  acc1: 90.6250 (74.0000)  acc5: 100.0000 (92.4980)  time: 0.1262  data: 0.0001  max mem: 17867
[09:33:58.649724] Test: Total time: 0:03:24 (0.1306 s / it)
[09:33:58.949157] * Acc@1 74.000 Acc@5 92.498 loss 1.058
[09:33:58.949323] Accuracy of the network on the 50000 test images: 74.0%
[09:33:58.949345] Max accuracy: 74.00%
[09:33:58.980211] log_dir: ./output_dir_cml_spikformer
[09:34:00.661188] Epoch: [130]  [   0/5004]  eta: 2:20:08  lr: 0.000343  loss: 2.8540 (2.8540)  time: 1.6803  data: 1.1653  max mem: 17867
[09:45:39.851096] Epoch: [130]  [2000/5004]  eta: 0:17:32  lr: 0.000340  loss: 2.8499 (2.7802)  time: 0.3484  data: 0.0002  max mem: 17867
[09:57:18.492407] Epoch: [130]  [4000/5004]  eta: 0:05:51  lr: 0.000337  loss: 2.5940 (2.7839)  time: 0.3498  data: 0.0002  max mem: 17867
[10:03:08.832971] Epoch: [130]  [5003/5004]  eta: 0:00:00  lr: 0.000335  loss: 2.8773 (2.7851)  time: 0.3453  data: 0.0011  max mem: 17867
[10:03:09.230955] Epoch: [130] Total time: 0:29:10 (0.3498 s / it)
[10:03:09.238888] Averaged stats: lr: 0.000335  loss: 2.8773 (2.7889)
[10:03:10.226173] Test:  [   0/1563]  eta: 0:25:35  loss: 0.5230 (0.5230)  acc1: 87.5000 (87.5000)  acc5: 96.8750 (96.8750)  time: 0.9824  data: 0.8029  max mem: 17867
[10:04:15.351743] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.8994 (0.8558)  acc1: 71.8750 (78.4057)  acc5: 96.8750 (95.3718)  time: 0.1299  data: 0.0002  max mem: 17867
[10:05:20.340641] Test:  [1000/1563]  eta: 0:01:13  loss: 0.8931 (0.9744)  acc1: 71.8750 (75.9709)  acc5: 93.7500 (93.6626)  time: 0.1299  data: 0.0002  max mem: 17867
[10:06:25.320469] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5650 (1.0608)  acc1: 87.5000 (74.0506)  acc5: 96.8750 (92.4550)  time: 0.1299  data: 0.0002  max mem: 17867
[10:06:33.304999] Test:  [1562/1563]  eta: 0:00:00  loss: 0.5578 (1.0616)  acc1: 90.6250 (73.9840)  acc5: 96.8750 (92.4640)  time: 0.1262  data: 0.0001  max mem: 17867
[10:06:33.386921] Test: Total time: 0:03:24 (0.1306 s / it)
[10:06:33.582220] * Acc@1 73.984 Acc@5 92.464 loss 1.062
[10:06:33.582360] Accuracy of the network on the 50000 test images: 74.0%
[10:06:33.582385] Max accuracy: 74.00%
[10:06:33.615371] log_dir: ./output_dir_cml_spikformer
[10:06:35.095537] Epoch: [131]  [   0/5004]  eta: 2:03:22  lr: 0.000335  loss: 2.8336 (2.8336)  time: 1.4794  data: 1.0477  max mem: 17867
[10:18:14.290236] Epoch: [131]  [2000/5004]  eta: 0:17:31  lr: 0.000331  loss: 2.7304 (2.7758)  time: 0.3529  data: 0.0002  max mem: 17867
[10:29:53.795194] Epoch: [131]  [4000/5004]  eta: 0:05:51  lr: 0.000328  loss: 2.7751 (2.7784)  time: 0.3464  data: 0.0002  max mem: 17867
[10:35:45.288870] Epoch: [131]  [5003/5004]  eta: 0:00:00  lr: 0.000326  loss: 2.7242 (2.7832)  time: 0.3442  data: 0.0014  max mem: 17867
[10:35:45.644004] Epoch: [131] Total time: 0:29:12 (0.3501 s / it)
[10:35:45.649622] Averaged stats: lr: 0.000326  loss: 2.7242 (2.7825)
[10:35:46.678912] Test:  [   0/1563]  eta: 0:26:42  loss: 0.5445 (0.5445)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.0256  data: 0.8851  max mem: 17867
[10:36:51.686132] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.8271 (0.8277)  acc1: 78.1250 (78.8922)  acc5: 96.8750 (95.6774)  time: 0.1300  data: 0.0002  max mem: 17867
[10:37:56.679915] Test:  [1000/1563]  eta: 0:01:13  loss: 1.3313 (0.9663)  acc1: 65.6250 (75.7149)  acc5: 90.6250 (93.7750)  time: 0.1298  data: 0.0002  max mem: 17867
[10:39:01.681762] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5238 (1.0543)  acc1: 87.5000 (73.7987)  acc5: 93.7500 (92.5050)  time: 0.1298  data: 0.0002  max mem: 17867
[10:39:09.661024] Test:  [1562/1563]  eta: 0:00:00  loss: 0.5599 (1.0575)  acc1: 87.5000 (73.7600)  acc5: 96.8750 (92.4700)  time: 0.1262  data: 0.0001  max mem: 17867
[10:39:09.736222] Test: Total time: 0:03:24 (0.1306 s / it)
[10:39:10.053897] * Acc@1 73.760 Acc@5 92.470 loss 1.057
[10:39:10.054141] Accuracy of the network on the 50000 test images: 73.8%
[10:39:10.054163] Max accuracy: 74.00%
[10:39:10.222666] log_dir: ./output_dir_cml_spikformer
[10:39:11.805217] Epoch: [132]  [   0/5004]  eta: 2:11:53  lr: 0.000326  loss: 2.7708 (2.7708)  time: 1.5815  data: 1.1914  max mem: 17867
[10:50:51.218402] Epoch: [132]  [2000/5004]  eta: 0:17:32  lr: 0.000323  loss: 2.7757 (2.7604)  time: 0.3511  data: 0.0002  max mem: 17867
[11:02:29.499601] Epoch: [132]  [4000/5004]  eta: 0:05:51  lr: 0.000319  loss: 2.7729 (2.7778)  time: 0.3473  data: 0.0002  max mem: 17867
[11:08:19.548525] Epoch: [132]  [5003/5004]  eta: 0:00:00  lr: 0.000318  loss: 2.6379 (2.7785)  time: 0.3463  data: 0.0011  max mem: 17867
[11:08:19.909053] Epoch: [132] Total time: 0:29:09 (0.3497 s / it)
[11:08:19.915670] Averaged stats: lr: 0.000318  loss: 2.6379 (2.7758)
[11:08:20.896196] Test:  [   0/1563]  eta: 0:25:27  loss: 0.3631 (0.3631)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 0.9771  data: 0.8070  max mem: 17867
[11:09:25.871670] Test:  [ 500/1563]  eta: 0:02:19  loss: 0.7725 (0.8190)  acc1: 81.2500 (78.6801)  acc5: 96.8750 (95.7959)  time: 0.1298  data: 0.0002  max mem: 17867
[11:10:30.830391] Test:  [1000/1563]  eta: 0:01:13  loss: 1.3098 (0.9679)  acc1: 68.7500 (75.4339)  acc5: 90.6250 (93.6813)  time: 0.1307  data: 0.0002  max mem: 17867
[11:11:35.772133] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5893 (1.0499)  acc1: 87.5000 (73.7196)  acc5: 93.7500 (92.5633)  time: 0.1298  data: 0.0002  max mem: 17867
[11:11:43.747726] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4533 (1.0504)  acc1: 90.6250 (73.7080)  acc5: 96.8750 (92.5460)  time: 0.1261  data: 0.0001  max mem: 17867
[11:11:43.808179] Test: Total time: 0:03:23 (0.1304 s / it)
[11:11:44.116475] * Acc@1 73.708 Acc@5 92.546 loss 1.050
[11:11:44.116641] Accuracy of the network on the 50000 test images: 73.7%
[11:11:44.116671] Max accuracy: 74.00%
[11:11:44.133918] log_dir: ./output_dir_cml_spikformer
[11:11:45.597225] Epoch: [133]  [   0/5004]  eta: 2:01:56  lr: 0.000318  loss: 2.2863 (2.2863)  time: 1.4621  data: 0.8936  max mem: 17867
[11:23:23.576185] Epoch: [133]  [2000/5004]  eta: 0:17:29  lr: 0.000314  loss: 2.6103 (2.7502)  time: 0.3447  data: 0.0002  max mem: 17867
[11:35:00.608992] Epoch: [133]  [4000/5004]  eta: 0:05:50  lr: 0.000311  loss: 2.7254 (2.7605)  time: 0.3508  data: 0.0002  max mem: 17867
[11:40:50.386404] Epoch: [133]  [5003/5004]  eta: 0:00:00  lr: 0.000309  loss: 2.7272 (2.7626)  time: 0.3430  data: 0.0006  max mem: 17867
[11:40:50.752934] Epoch: [133] Total time: 0:29:06 (0.3490 s / it)
[11:40:50.758778] Averaged stats: lr: 0.000309  loss: 2.7272 (2.7700)
[11:40:52.007857] Test:  [   0/1563]  eta: 0:32:26  loss: 0.3791 (0.3791)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.2457  data: 1.0668  max mem: 17867
[11:41:57.025348] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.9508 (0.8373)  acc1: 75.0000 (78.9172)  acc5: 96.8750 (95.5215)  time: 0.1300  data: 0.0002  max mem: 17867
[11:43:02.005134] Test:  [1000/1563]  eta: 0:01:13  loss: 1.4094 (0.9842)  acc1: 65.6250 (75.7493)  acc5: 90.6250 (93.5377)  time: 0.1299  data: 0.0002  max mem: 17867
[11:44:06.989309] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5708 (1.0666)  acc1: 84.3750 (74.0548)  acc5: 96.8750 (92.3822)  time: 0.1299  data: 0.0002  max mem: 17867
[11:44:14.974824] Test:  [1562/1563]  eta: 0:00:00  loss: 0.5768 (1.0665)  acc1: 87.5000 (74.0480)  acc5: 96.8750 (92.4040)  time: 0.1262  data: 0.0001  max mem: 17867
[11:44:15.037047] Test: Total time: 0:03:24 (0.1307 s / it)
[11:44:15.165729] * Acc@1 74.048 Acc@5 92.404 loss 1.067
[11:44:15.165863] Accuracy of the network on the 50000 test images: 74.0%
[11:44:15.165883] Max accuracy: 74.05%
[11:44:15.199578] log_dir: ./output_dir_cml_spikformer
[11:44:16.709084] Epoch: [134]  [   0/5004]  eta: 2:05:47  lr: 0.000309  loss: 2.9030 (2.9030)  time: 1.5083  data: 1.0479  max mem: 17867
[11:55:54.730970] Epoch: [134]  [2000/5004]  eta: 0:17:30  lr: 0.000306  loss: 2.7302 (2.7557)  time: 0.3523  data: 0.0002  max mem: 17867
[12:07:32.535689] Epoch: [134]  [4000/5004]  eta: 0:05:50  lr: 0.000302  loss: 2.9107 (2.7584)  time: 0.3478  data: 0.0002  max mem: 17867
[12:13:22.606938] Epoch: [134]  [5003/5004]  eta: 0:00:00  lr: 0.000301  loss: 2.6966 (2.7573)  time: 0.3468  data: 0.0011  max mem: 17867
[12:13:22.987939] Epoch: [134] Total time: 0:29:07 (0.3493 s / it)
[12:13:22.994094] Averaged stats: lr: 0.000301  loss: 2.6966 (2.7586)
[12:13:24.099494] Test:  [   0/1563]  eta: 0:28:42  loss: 0.4780 (0.4780)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.1020  data: 0.9093  max mem: 17867
[12:14:29.089804] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.8136 (0.8430)  acc1: 75.0000 (79.0357)  acc5: 96.8750 (95.5776)  time: 0.1300  data: 0.0002  max mem: 17867
[12:15:34.954599] Test:  [1000/1563]  eta: 0:01:14  loss: 1.1795 (0.9788)  acc1: 71.8750 (75.9085)  acc5: 93.7500 (93.7656)  time: 0.1310  data: 0.0002  max mem: 17867
[12:16:40.418709] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5617 (1.0565)  acc1: 84.3750 (74.1131)  acc5: 96.8750 (92.6653)  time: 0.1309  data: 0.0002  max mem: 17867
[12:16:48.409692] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4689 (1.0560)  acc1: 87.5000 (74.0980)  acc5: 96.8750 (92.6780)  time: 0.1262  data: 0.0001  max mem: 17867
[12:16:48.479313] Test: Total time: 0:03:25 (0.1315 s / it)
[12:16:48.490813] * Acc@1 74.098 Acc@5 92.678 loss 1.056
[12:16:48.490961] Accuracy of the network on the 50000 test images: 74.1%
[12:16:48.490983] Max accuracy: 74.10%
[12:16:48.497275] log_dir: ./output_dir_cml_spikformer
[12:16:49.942326] Epoch: [135]  [   0/5004]  eta: 2:00:24  lr: 0.000301  loss: 2.3092 (2.3092)  time: 1.4438  data: 0.8658  max mem: 17867
[12:28:28.729774] Epoch: [135]  [2000/5004]  eta: 0:17:31  lr: 0.000297  loss: 2.7417 (2.7606)  time: 0.3465  data: 0.0002  max mem: 17867
[12:40:06.141285] Epoch: [135]  [4000/5004]  eta: 0:05:50  lr: 0.000294  loss: 2.6516 (2.7650)  time: 0.3461  data: 0.0002  max mem: 17867
[12:45:55.823997] Epoch: [135]  [5003/5004]  eta: 0:00:00  lr: 0.000292  loss: 2.8761 (2.7645)  time: 0.3457  data: 0.0006  max mem: 17867
[12:45:56.192427] Epoch: [135] Total time: 0:29:07 (0.3493 s / it)
[12:45:56.196454] Averaged stats: lr: 0.000292  loss: 2.8761 (2.7552)
[12:45:57.240413] Test:  [   0/1563]  eta: 0:27:05  loss: 0.4346 (0.4346)  acc1: 90.6250 (90.6250)  acc5: 96.8750 (96.8750)  time: 1.0401  data: 0.8790  max mem: 17867
[12:47:02.446944] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.8972 (0.8381)  acc1: 75.0000 (78.8797)  acc5: 96.8750 (95.6587)  time: 0.1300  data: 0.0002  max mem: 17867
[12:48:07.441222] Test:  [1000/1563]  eta: 0:01:13  loss: 1.2138 (0.9637)  acc1: 65.6250 (76.1332)  acc5: 90.6250 (93.8561)  time: 0.1300  data: 0.0002  max mem: 17867
[12:49:12.423829] Test:  [1500/1563]  eta: 0:00:08  loss: 0.6476 (1.0500)  acc1: 84.3750 (74.2692)  acc5: 96.8750 (92.6591)  time: 0.1299  data: 0.0002  max mem: 17867
[12:49:20.465115] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4592 (1.0486)  acc1: 87.5000 (74.2760)  acc5: 96.8750 (92.6940)  time: 0.1291  data: 0.0001  max mem: 17867
[12:49:20.608188] Test: Total time: 0:03:24 (0.1308 s / it)
[12:49:20.615161] * Acc@1 74.276 Acc@5 92.694 loss 1.049
[12:49:20.615285] Accuracy of the network on the 50000 test images: 74.3%
[12:49:20.615310] Max accuracy: 74.28%
[12:49:20.622511] log_dir: ./output_dir_cml_spikformer
[12:49:22.165340] Epoch: [136]  [   0/5004]  eta: 2:08:37  lr: 0.000292  loss: 2.4926 (2.4926)  time: 1.5422  data: 1.0425  max mem: 17867
[13:00:59.738997] Epoch: [136]  [2000/5004]  eta: 0:17:29  lr: 0.000289  loss: 2.5610 (2.7396)  time: 0.3507  data: 0.0002  max mem: 17867
[13:12:36.310868] Epoch: [136]  [4000/5004]  eta: 0:05:50  lr: 0.000286  loss: 2.8004 (2.7464)  time: 0.3460  data: 0.0002  max mem: 17867
[13:18:26.630175] Epoch: [136]  [5003/5004]  eta: 0:00:00  lr: 0.000284  loss: 2.6099 (2.7488)  time: 0.3439  data: 0.0011  max mem: 17867
[13:18:26.945722] Epoch: [136] Total time: 0:29:06 (0.3490 s / it)
[13:18:26.954663] Averaged stats: lr: 0.000284  loss: 2.6099 (2.7457)
[13:18:28.063574] Test:  [   0/1563]  eta: 0:28:45  loss: 0.3778 (0.3778)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.1041  data: 0.9107  max mem: 17867
[13:19:33.077204] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.0080 (0.8451)  acc1: 75.0000 (78.5367)  acc5: 96.8750 (95.7335)  time: 0.1299  data: 0.0002  max mem: 17867
[13:20:38.211924] Test:  [1000/1563]  eta: 0:01:13  loss: 1.1230 (0.9666)  acc1: 65.6250 (75.9522)  acc5: 93.7500 (93.9217)  time: 0.1299  data: 0.0002  max mem: 17867
[13:21:43.175161] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5992 (1.0488)  acc1: 84.3750 (74.2692)  acc5: 96.8750 (92.5987)  time: 0.1299  data: 0.0002  max mem: 17867
[13:21:51.156061] Test:  [1562/1563]  eta: 0:00:00  loss: 0.3389 (1.0511)  acc1: 90.6250 (74.1940)  acc5: 96.8750 (92.5960)  time: 0.1262  data: 0.0001  max mem: 17867
[13:21:51.253112] Test: Total time: 0:03:24 (0.1307 s / it)
[13:21:51.256099] * Acc@1 74.194 Acc@5 92.596 loss 1.051
[13:21:51.256260] Accuracy of the network on the 50000 test images: 74.2%
[13:21:51.256283] Max accuracy: 74.28%
[13:21:51.263826] log_dir: ./output_dir_cml_spikformer
[13:21:52.853496] Epoch: [137]  [   0/5004]  eta: 2:12:31  lr: 0.000284  loss: 2.6487 (2.6487)  time: 1.5891  data: 0.9308  max mem: 17867
[13:33:32.029757] Epoch: [137]  [2000/5004]  eta: 0:17:31  lr: 0.000281  loss: 2.7532 (2.7400)  time: 0.3473  data: 0.0002  max mem: 17867
[13:45:09.493787] Epoch: [137]  [4000/5004]  eta: 0:05:50  lr: 0.000278  loss: 2.6886 (2.7370)  time: 0.3484  data: 0.0002  max mem: 17867
[13:50:59.499023] Epoch: [137]  [5003/5004]  eta: 0:00:00  lr: 0.000276  loss: 2.6245 (2.7356)  time: 0.3476  data: 0.0012  max mem: 17867
[13:50:59.964420] Epoch: [137] Total time: 0:29:08 (0.3495 s / it)
[13:50:59.972693] Averaged stats: lr: 0.000276  loss: 2.6245 (2.7390)
[13:51:01.185050] Test:  [   0/1563]  eta: 0:31:29  loss: 0.4241 (0.4241)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.2087  data: 1.0631  max mem: 17867
[13:52:06.202597] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.6813 (0.8151)  acc1: 78.1250 (79.2789)  acc5: 93.7500 (95.8458)  time: 0.1300  data: 0.0002  max mem: 17867
[13:53:11.232514] Test:  [1000/1563]  eta: 0:01:13  loss: 1.1992 (0.9476)  acc1: 68.7500 (76.2831)  acc5: 93.7500 (93.9560)  time: 0.1300  data: 0.0002  max mem: 17867
[13:54:16.223136] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5314 (1.0335)  acc1: 84.3750 (74.2859)  acc5: 96.8750 (92.7236)  time: 0.1299  data: 0.0002  max mem: 17867
[13:54:24.210170] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4560 (1.0364)  acc1: 87.5000 (74.2080)  acc5: 100.0000 (92.7180)  time: 0.1262  data: 0.0001  max mem: 17867
[13:54:24.305091] Test: Total time: 0:03:24 (0.1307 s / it)
[13:54:24.405893] * Acc@1 74.208 Acc@5 92.718 loss 1.036
[13:54:24.406194] Accuracy of the network on the 50000 test images: 74.2%
[13:54:24.406219] Max accuracy: 74.28%
[13:54:24.457380] log_dir: ./output_dir_cml_spikformer
[13:54:26.160466] Epoch: [138]  [   0/5004]  eta: 2:21:59  lr: 0.000276  loss: 3.2226 (3.2226)  time: 1.7025  data: 1.1650  max mem: 17867
[14:06:05.388633] Epoch: [138]  [2000/5004]  eta: 0:17:32  lr: 0.000273  loss: 2.7964 (2.7166)  time: 0.3506  data: 0.0002  max mem: 17867
[14:17:43.856997] Epoch: [138]  [4000/5004]  eta: 0:05:51  lr: 0.000270  loss: 2.7103 (2.7246)  time: 0.3501  data: 0.0004  max mem: 17867
[14:23:34.184222] Epoch: [138]  [5003/5004]  eta: 0:00:00  lr: 0.000268  loss: 2.6764 (2.7252)  time: 0.3456  data: 0.0007  max mem: 17867
[14:23:34.512206] Epoch: [138] Total time: 0:29:10 (0.3497 s / it)
[14:23:34.514058] Averaged stats: lr: 0.000268  loss: 2.6764 (2.7280)
[14:23:35.456018] Test:  [   0/1563]  eta: 0:24:26  loss: 0.5871 (0.5871)  acc1: 90.6250 (90.6250)  acc5: 96.8750 (96.8750)  time: 0.9383  data: 0.7992  max mem: 17867
[14:24:40.466200] Test:  [ 500/1563]  eta: 0:02:19  loss: 0.6534 (0.8262)  acc1: 81.2500 (79.4349)  acc5: 96.8750 (95.6712)  time: 0.1300  data: 0.0002  max mem: 17867
[14:25:45.541374] Test:  [1000/1563]  eta: 0:01:13  loss: 1.1562 (0.9527)  acc1: 71.8750 (76.5984)  acc5: 90.6250 (93.8780)  time: 0.1301  data: 0.0002  max mem: 17867
[14:26:50.540220] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5215 (1.0314)  acc1: 87.5000 (74.7314)  acc5: 96.8750 (92.8110)  time: 0.1300  data: 0.0002  max mem: 17867
[14:26:58.548394] Test:  [1562/1563]  eta: 0:00:00  loss: 0.3943 (1.0312)  acc1: 90.6250 (74.7000)  acc5: 96.8750 (92.8340)  time: 0.1269  data: 0.0001  max mem: 17867
[14:26:58.631991] Test: Total time: 0:03:24 (0.1306 s / it)
[14:26:58.841761] * Acc@1 74.700 Acc@5 92.834 loss 1.031
[14:26:58.841960] Accuracy of the network on the 50000 test images: 74.7%
[14:26:58.841987] Max accuracy: 74.70%
[14:26:58.848905] log_dir: ./output_dir_cml_spikformer
[14:27:00.522258] Epoch: [139]  [   0/5004]  eta: 2:19:27  lr: 0.000268  loss: 2.4417 (2.4417)  time: 1.6723  data: 1.0885  max mem: 17867
[14:38:39.371966] Epoch: [139]  [2000/5004]  eta: 0:17:31  lr: 0.000265  loss: 2.5685 (2.7140)  time: 0.3499  data: 0.0002  max mem: 17867
[14:50:17.914919] Epoch: [139]  [4000/5004]  eta: 0:05:51  lr: 0.000262  loss: 2.6048 (2.7202)  time: 0.3473  data: 0.0002  max mem: 17867
[14:56:08.083949] Epoch: [139]  [5003/5004]  eta: 0:00:00  lr: 0.000260  loss: 2.6836 (2.7235)  time: 0.3471  data: 0.0011  max mem: 17867
[14:56:08.454228] Epoch: [139] Total time: 0:29:09 (0.3496 s / it)
[14:56:08.454972] Averaged stats: lr: 0.000260  loss: 2.6836 (2.7211)
[14:56:09.435167] Test:  [   0/1563]  eta: 0:25:26  loss: 0.3662 (0.3662)  acc1: 90.6250 (90.6250)  acc5: 96.8750 (96.8750)  time: 0.9766  data: 0.8368  max mem: 17867
[14:57:14.495969] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.8706 (0.8145)  acc1: 78.1250 (79.3538)  acc5: 96.8750 (95.8333)  time: 0.1300  data: 0.0002  max mem: 17867
[14:58:19.529151] Test:  [1000/1563]  eta: 0:01:13  loss: 1.0550 (0.9458)  acc1: 68.7500 (76.5453)  acc5: 93.7500 (93.9904)  time: 0.1308  data: 0.0002  max mem: 17867
[14:59:24.526354] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5159 (1.0247)  acc1: 87.5000 (74.8709)  acc5: 96.8750 (92.9422)  time: 0.1300  data: 0.0002  max mem: 17867
[14:59:32.513382] Test:  [1562/1563]  eta: 0:00:00  loss: 0.5208 (1.0265)  acc1: 90.6250 (74.8100)  acc5: 96.8750 (92.9380)  time: 0.1262  data: 0.0001  max mem: 17867
[14:59:32.590006] Test: Total time: 0:03:24 (0.1306 s / it)
[14:59:32.857303] * Acc@1 74.810 Acc@5 92.938 loss 1.027
[14:59:32.857441] Accuracy of the network on the 50000 test images: 74.8%
[14:59:32.857459] Max accuracy: 74.81%
[14:59:32.876285] log_dir: ./output_dir_cml_spikformer
[14:59:34.396032] Epoch: [140]  [   0/5004]  eta: 2:06:38  lr: 0.000260  loss: 2.6569 (2.6569)  time: 1.5185  data: 1.1785  max mem: 17867
[15:11:15.200008] Epoch: [140]  [2000/5004]  eta: 0:17:34  lr: 0.000257  loss: 2.5819 (2.7127)  time: 0.3507  data: 0.0002  max mem: 17867
[15:22:54.921949] Epoch: [140]  [4000/5004]  eta: 0:05:51  lr: 0.000254  loss: 2.7567 (2.7141)  time: 0.3492  data: 0.0002  max mem: 17867
[15:28:45.376783] Epoch: [140]  [5003/5004]  eta: 0:00:00  lr: 0.000252  loss: 2.7033 (2.7164)  time: 0.3464  data: 0.0011  max mem: 17867
[15:28:45.755295] Epoch: [140] Total time: 0:29:12 (0.3503 s / it)
[15:28:45.760130] Averaged stats: lr: 0.000252  loss: 2.7033 (2.7124)
[15:28:46.851471] Test:  [   0/1563]  eta: 0:28:19  loss: 0.4009 (0.4009)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.0875  data: 0.9010  max mem: 17867
[15:29:51.914333] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.7369 (0.7969)  acc1: 75.0000 (79.7717)  acc5: 96.8750 (96.1390)  time: 0.1299  data: 0.0002  max mem: 17867
[15:30:56.906309] Test:  [1000/1563]  eta: 0:01:13  loss: 1.2332 (0.9330)  acc1: 71.8750 (76.7264)  acc5: 93.7500 (94.2714)  time: 0.1299  data: 0.0002  max mem: 17867
[15:32:01.892121] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5314 (1.0127)  acc1: 84.3750 (75.0999)  acc5: 96.8750 (93.1566)  time: 0.1299  data: 0.0002  max mem: 17867
[15:32:09.874118] Test:  [1562/1563]  eta: 0:00:00  loss: 0.5313 (1.0135)  acc1: 90.6250 (75.0560)  acc5: 96.8750 (93.1580)  time: 0.1262  data: 0.0001  max mem: 17867
[15:32:09.979177] Test: Total time: 0:03:24 (0.1307 s / it)
[15:32:10.062756] * Acc@1 75.056 Acc@5 93.158 loss 1.013
[15:32:10.062916] Accuracy of the network on the 50000 test images: 75.1%
[15:32:10.062941] Max accuracy: 75.06%
[15:32:10.197050] log_dir: ./output_dir_cml_spikformer
[15:32:11.740261] Epoch: [141]  [   0/5004]  eta: 2:08:38  lr: 0.000252  loss: 2.5950 (2.5950)  time: 1.5426  data: 1.1946  max mem: 17867
[15:43:51.982504] Epoch: [141]  [2000/5004]  eta: 0:17:33  lr: 0.000249  loss: 2.8139 (2.6896)  time: 0.3490  data: 0.0002  max mem: 17867
[15:55:30.838413] Epoch: [141]  [4000/5004]  eta: 0:05:51  lr: 0.000246  loss: 2.7687 (2.7021)  time: 0.3544  data: 0.0002  max mem: 17867
[16:01:21.623747] Epoch: [141]  [5003/5004]  eta: 0:00:00  lr: 0.000244  loss: 2.6490 (2.7037)  time: 0.3481  data: 0.0013  max mem: 17867
[16:01:21.996452] Epoch: [141] Total time: 0:29:11 (0.3501 s / it)
[16:01:22.004010] Averaged stats: lr: 0.000244  loss: 2.6490 (2.7077)
[16:01:23.115483] Test:  [   0/1563]  eta: 0:28:51  loss: 0.2440 (0.2440)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.1080  data: 0.9649  max mem: 17867
[16:02:28.143287] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.7356 (0.8094)  acc1: 78.1250 (79.5097)  acc5: 96.8750 (95.9643)  time: 0.1300  data: 0.0002  max mem: 17867
[16:03:33.269014] Test:  [1000/1563]  eta: 0:01:13  loss: 1.1836 (0.9506)  acc1: 71.8750 (76.3580)  acc5: 93.7500 (94.0653)  time: 0.1299  data: 0.0002  max mem: 17867
[16:04:38.263281] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5182 (1.0265)  acc1: 87.5000 (74.7127)  acc5: 93.7500 (93.0192)  time: 0.1299  data: 0.0002  max mem: 17867
[16:04:46.248354] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4668 (1.0261)  acc1: 90.6250 (74.7380)  acc5: 96.8750 (93.0260)  time: 0.1262  data: 0.0001  max mem: 17867
[16:04:46.366383] Test: Total time: 0:03:24 (0.1307 s / it)
[16:04:46.386153] * Acc@1 74.738 Acc@5 93.026 loss 1.026
[16:04:46.386293] Accuracy of the network on the 50000 test images: 74.7%
[16:04:46.386315] Max accuracy: 75.06%
[16:04:46.399432] log_dir: ./output_dir_cml_spikformer
[16:04:48.044587] Epoch: [142]  [   0/5004]  eta: 2:17:07  lr: 0.000244  loss: 2.6248 (2.6248)  time: 1.6442  data: 1.2883  max mem: 17867
[16:16:28.693239] Epoch: [142]  [2000/5004]  eta: 0:17:34  lr: 0.000241  loss: 2.6824 (2.6993)  time: 0.3560  data: 0.0002  max mem: 17867
[16:28:08.023832] Epoch: [142]  [4000/5004]  eta: 0:05:51  lr: 0.000238  loss: 2.6806 (2.6991)  time: 0.3478  data: 0.0002  max mem: 17867
[16:33:59.011143] Epoch: [142]  [5003/5004]  eta: 0:00:00  lr: 0.000237  loss: 2.6469 (2.7009)  time: 0.3477  data: 0.0011  max mem: 17867
[16:33:59.386912] Epoch: [142] Total time: 0:29:12 (0.3503 s / it)
[16:33:59.387575] Averaged stats: lr: 0.000237  loss: 2.6469 (2.6987)
[16:34:00.421890] Test:  [   0/1563]  eta: 0:26:50  loss: 0.3253 (0.3253)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.0305  data: 0.8929  max mem: 17867
[16:35:05.438578] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.6653 (0.7996)  acc1: 78.1250 (79.9900)  acc5: 96.8750 (95.9269)  time: 0.1301  data: 0.0002  max mem: 17867
[16:36:10.437630] Test:  [1000/1563]  eta: 0:01:13  loss: 1.3396 (0.9373)  acc1: 59.3750 (76.7326)  acc5: 93.7500 (94.1933)  time: 0.1299  data: 0.0002  max mem: 17867
[16:37:15.487139] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5035 (1.0154)  acc1: 87.5000 (75.0000)  acc5: 96.8750 (93.0546)  time: 0.1299  data: 0.0002  max mem: 17867
[16:37:23.471109] Test:  [1562/1563]  eta: 0:00:00  loss: 0.3983 (1.0166)  acc1: 90.6250 (74.9240)  acc5: 100.0000 (93.0580)  time: 0.1263  data: 0.0001  max mem: 17867
[16:37:23.533102] Test: Total time: 0:03:24 (0.1306 s / it)
[16:37:23.950408] * Acc@1 74.924 Acc@5 93.058 loss 1.017
[16:37:23.950587] Accuracy of the network on the 50000 test images: 74.9%
[16:37:23.950609] Max accuracy: 75.06%
[16:37:23.981035] log_dir: ./output_dir_cml_spikformer
[16:37:25.455755] Epoch: [143]  [   0/5004]  eta: 2:02:53  lr: 0.000237  loss: 2.4679 (2.4679)  time: 1.4735  data: 1.0448  max mem: 17867
[16:49:05.253902] Epoch: [143]  [2000/5004]  eta: 0:17:32  lr: 0.000233  loss: 2.8529 (2.6873)  time: 0.3491  data: 0.0002  max mem: 17867
[17:00:44.766384] Epoch: [143]  [4000/5004]  eta: 0:05:51  lr: 0.000230  loss: 2.7109 (2.6982)  time: 0.3505  data: 0.0002  max mem: 17867
[17:06:35.739075] Epoch: [143]  [5003/5004]  eta: 0:00:00  lr: 0.000229  loss: 2.7441 (2.6992)  time: 0.3448  data: 0.0011  max mem: 17867
[17:06:36.140654] Epoch: [143] Total time: 0:29:12 (0.3502 s / it)
[17:06:36.143110] Averaged stats: lr: 0.000229  loss: 2.7441 (2.6925)
[17:06:37.543814] Test:  [   0/1563]  eta: 0:36:21  loss: 0.4005 (0.4005)  acc1: 90.6250 (90.6250)  acc5: 96.8750 (96.8750)  time: 1.3959  data: 1.2518  max mem: 17867
[17:07:42.607266] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.9269 (0.7955)  acc1: 75.0000 (80.0399)  acc5: 93.7500 (96.1390)  time: 0.1299  data: 0.0002  max mem: 17867
[17:08:47.631953] Test:  [1000/1563]  eta: 0:01:13  loss: 1.2484 (0.9286)  acc1: 71.8750 (77.0386)  acc5: 93.7500 (94.2776)  time: 0.1309  data: 0.0002  max mem: 17867
[17:09:52.635395] Test:  [1500/1563]  eta: 0:00:08  loss: 0.6388 (1.0027)  acc1: 87.5000 (75.3040)  acc5: 96.8750 (93.2607)  time: 0.1300  data: 0.0002  max mem: 17867
[17:10:00.621400] Test:  [1562/1563]  eta: 0:00:00  loss: 0.3937 (1.0027)  acc1: 90.6250 (75.2960)  acc5: 96.8750 (93.2720)  time: 0.1262  data: 0.0001  max mem: 17867
[17:10:00.711321] Test: Total time: 0:03:24 (0.1309 s / it)
[17:10:00.802492] * Acc@1 75.296 Acc@5 93.272 loss 1.003
[17:10:00.802631] Accuracy of the network on the 50000 test images: 75.3%
[17:10:00.802657] Max accuracy: 75.30%
[17:10:00.830047] log_dir: ./output_dir_cml_spikformer
[17:10:02.335191] Epoch: [144]  [   0/5004]  eta: 2:05:28  lr: 0.000229  loss: 1.9735 (1.9735)  time: 1.5045  data: 1.0899  max mem: 17867
[17:21:43.610944] Epoch: [144]  [2000/5004]  eta: 0:17:35  lr: 0.000226  loss: 2.6725 (2.6778)  time: 0.3555  data: 0.0002  max mem: 17867
[17:33:23.463489] Epoch: [144]  [4000/5004]  eta: 0:05:51  lr: 0.000223  loss: 2.5895 (2.6776)  time: 0.3447  data: 0.0002  max mem: 17867
[17:39:14.651708] Epoch: [144]  [5003/5004]  eta: 0:00:00  lr: 0.000221  loss: 2.7492 (2.6815)  time: 0.3450  data: 0.0011  max mem: 17867
[17:39:15.047516] Epoch: [144] Total time: 0:29:14 (0.3506 s / it)
[17:39:15.049310] Averaged stats: lr: 0.000221  loss: 2.7492 (2.6816)
[17:39:16.122243] Test:  [   0/1563]  eta: 0:27:49  loss: 0.5166 (0.5166)  acc1: 87.5000 (87.5000)  acc5: 96.8750 (96.8750)  time: 1.0681  data: 0.9308  max mem: 17867
[17:40:21.127827] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.8890 (0.7939)  acc1: 75.0000 (79.5035)  acc5: 96.8750 (96.0828)  time: 0.1299  data: 0.0002  max mem: 17867
[17:41:26.115972] Test:  [1000/1563]  eta: 0:01:13  loss: 1.4646 (0.9281)  acc1: 56.2500 (76.8638)  acc5: 90.6250 (94.1496)  time: 0.1299  data: 0.0002  max mem: 17867
[17:42:31.100948] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5339 (1.0054)  acc1: 87.5000 (75.1041)  acc5: 96.8750 (93.2441)  time: 0.1299  data: 0.0002  max mem: 17867
[17:42:39.091981] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4282 (1.0074)  acc1: 90.6250 (75.0420)  acc5: 96.8750 (93.2360)  time: 0.1265  data: 0.0001  max mem: 17867
[17:42:39.239998] Test: Total time: 0:03:24 (0.1306 s / it)
[17:42:39.397511] * Acc@1 75.042 Acc@5 93.236 loss 1.007
[17:42:39.397666] Accuracy of the network on the 50000 test images: 75.0%
[17:42:39.397687] Max accuracy: 75.30%
[17:42:39.571710] log_dir: ./output_dir_cml_spikformer
[17:42:41.159392] Epoch: [145]  [   0/5004]  eta: 2:12:20  lr: 0.000221  loss: 2.3548 (2.3548)  time: 1.5869  data: 1.2574  max mem: 17867
[17:54:22.343488] Epoch: [145]  [2000/5004]  eta: 0:17:34  lr: 0.000218  loss: 2.5626 (2.6703)  time: 0.3469  data: 0.0002  max mem: 17867
[18:06:02.610043] Epoch: [145]  [4000/5004]  eta: 0:05:52  lr: 0.000215  loss: 2.7011 (2.6728)  time: 0.3506  data: 0.0002  max mem: 17867
[18:11:53.260576] Epoch: [145]  [5003/5004]  eta: 0:00:00  lr: 0.000214  loss: 2.6161 (2.6743)  time: 0.3462  data: 0.0006  max mem: 17867
[18:11:53.643416] Epoch: [145] Total time: 0:29:14 (0.3505 s / it)
[18:11:53.652388] Averaged stats: lr: 0.000214  loss: 2.6161 (2.6755)
[18:11:55.032583] Test:  [   0/1563]  eta: 0:35:51  loss: 0.2289 (0.2289)  acc1: 96.8750 (96.8750)  acc5: 100.0000 (100.0000)  time: 1.3768  data: 1.2389  max mem: 17867
[18:13:00.044252] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.6980 (0.8008)  acc1: 78.1250 (80.1272)  acc5: 96.8750 (96.1514)  time: 0.1299  data: 0.0002  max mem: 17867
[18:14:05.036672] Test:  [1000/1563]  eta: 0:01:13  loss: 1.1224 (0.9370)  acc1: 68.7500 (76.9637)  acc5: 93.7500 (94.2339)  time: 0.1299  data: 0.0002  max mem: 17867
[18:15:10.050952] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5335 (1.0143)  acc1: 87.5000 (75.2269)  acc5: 96.8750 (93.1691)  time: 0.1299  data: 0.0002  max mem: 17867
[18:15:18.060761] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4560 (1.0153)  acc1: 87.5000 (75.1920)  acc5: 96.8750 (93.1840)  time: 0.1265  data: 0.0001  max mem: 17867
[18:15:18.141799] Test: Total time: 0:03:24 (0.1308 s / it)
[18:15:18.320559] * Acc@1 75.192 Acc@5 93.184 loss 1.015
[18:15:18.320703] Accuracy of the network on the 50000 test images: 75.2%
[18:15:18.320724] Max accuracy: 75.30%
[18:15:18.335940] log_dir: ./output_dir_cml_spikformer
[18:15:19.899031] Epoch: [146]  [   0/5004]  eta: 2:10:18  lr: 0.000214  loss: 2.3287 (2.3287)  time: 1.5624  data: 1.1848  max mem: 17867
[18:27:01.093573] Epoch: [146]  [2000/5004]  eta: 0:17:34  lr: 0.000211  loss: 2.5709 (2.6611)  time: 0.3505  data: 0.0007  max mem: 17867
[18:38:41.769390] Epoch: [146]  [4000/5004]  eta: 0:05:52  lr: 0.000208  loss: 2.7549 (2.6597)  time: 0.3456  data: 0.0002  max mem: 17867
[18:44:32.837223] Epoch: [146]  [5003/5004]  eta: 0:00:00  lr: 0.000207  loss: 2.6238 (2.6637)  time: 0.3444  data: 0.0011  max mem: 17867
[18:44:33.307211] Epoch: [146] Total time: 0:29:14 (0.3507 s / it)
[18:44:33.307922] Averaged stats: lr: 0.000207  loss: 2.6238 (2.6713)
[18:44:34.386164] Test:  [   0/1563]  eta: 0:27:59  loss: 0.4819 (0.4819)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.0747  data: 0.9298  max mem: 17867
[18:45:39.403549] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.9108 (0.7850)  acc1: 78.1250 (79.8528)  acc5: 93.7500 (96.1577)  time: 0.1299  data: 0.0002  max mem: 17867
[18:46:44.405416] Test:  [1000/1563]  eta: 0:01:13  loss: 1.0811 (0.9228)  acc1: 71.8750 (76.7826)  acc5: 93.7500 (94.2308)  time: 0.1299  data: 0.0002  max mem: 17867
[18:47:49.365056] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5537 (1.0068)  acc1: 87.5000 (75.0187)  acc5: 100.0000 (93.1629)  time: 0.1299  data: 0.0002  max mem: 17867
[18:47:57.356376] Test:  [1562/1563]  eta: 0:00:00  loss: 0.3803 (1.0045)  acc1: 93.7500 (75.0320)  acc5: 96.8750 (93.2040)  time: 0.1262  data: 0.0001  max mem: 17867
[18:47:57.421436] Test: Total time: 0:03:24 (0.1306 s / it)
[18:47:57.897406] * Acc@1 75.032 Acc@5 93.204 loss 1.005
[18:47:57.897544] Accuracy of the network on the 50000 test images: 75.0%
[18:47:57.897566] Max accuracy: 75.30%
[18:47:57.942327] log_dir: ./output_dir_cml_spikformer
[18:47:59.588807] Epoch: [147]  [   0/5004]  eta: 2:17:13  lr: 0.000207  loss: 2.7353 (2.7353)  time: 1.6453  data: 1.2480  max mem: 17867
[18:59:40.330725] Epoch: [147]  [2000/5004]  eta: 0:17:34  lr: 0.000204  loss: 2.5956 (2.6438)  time: 0.3479  data: 0.0002  max mem: 17867
[19:11:20.292713] Epoch: [147]  [4000/5004]  eta: 0:05:51  lr: 0.000201  loss: 2.7359 (2.6495)  time: 0.3484  data: 0.0002  max mem: 17867
[19:17:11.400958] Epoch: [147]  [5003/5004]  eta: 0:00:00  lr: 0.000199  loss: 2.5780 (2.6554)  time: 0.3437  data: 0.0006  max mem: 17867
[19:17:11.834901] Epoch: [147] Total time: 0:29:13 (0.3505 s / it)
[19:17:11.836059] Averaged stats: lr: 0.000199  loss: 2.5780 (2.6596)
[19:17:13.251308] Test:  [   0/1563]  eta: 0:36:46  loss: 0.3922 (0.3922)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.4119  data: 1.2758  max mem: 17867
[19:18:18.303382] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.7952 (0.8110)  acc1: 78.1250 (79.6033)  acc5: 96.8750 (95.5838)  time: 0.1300  data: 0.0002  max mem: 17867
[19:19:23.263644] Test:  [1000/1563]  eta: 0:01:13  loss: 1.1961 (0.9196)  acc1: 62.5000 (77.1916)  acc5: 93.7500 (94.2714)  time: 0.1298  data: 0.0002  max mem: 17867
[19:20:28.256362] Test:  [1500/1563]  eta: 0:00:08  loss: 0.4488 (0.9952)  acc1: 87.5000 (75.4122)  acc5: 100.0000 (93.3440)  time: 0.1298  data: 0.0002  max mem: 17867
[19:20:36.236677] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4646 (0.9988)  acc1: 90.6250 (75.3100)  acc5: 96.8750 (93.3240)  time: 0.1261  data: 0.0001  max mem: 17867
[19:20:36.330809] Test: Total time: 0:03:24 (0.1308 s / it)
[19:20:36.491349] * Acc@1 75.310 Acc@5 93.324 loss 0.999
[19:20:36.491539] Accuracy of the network on the 50000 test images: 75.3%
[19:20:36.491564] Max accuracy: 75.31%
[19:20:36.498767] log_dir: ./output_dir_cml_spikformer
[19:20:38.323492] Epoch: [148]  [   0/5004]  eta: 2:32:08  lr: 0.000199  loss: 2.8543 (2.8543)  time: 1.8242  data: 1.0317  max mem: 17867
[19:32:17.254485] Epoch: [148]  [2000/5004]  eta: 0:17:31  lr: 0.000196  loss: 2.4743 (2.6455)  time: 0.3518  data: 0.0002  max mem: 17867
[19:43:55.933885] Epoch: [148]  [4000/5004]  eta: 0:05:51  lr: 0.000194  loss: 2.5832 (2.6532)  time: 0.3480  data: 0.0002  max mem: 17867
[19:49:46.113498] Epoch: [148]  [5003/5004]  eta: 0:00:00  lr: 0.000192  loss: 2.7203 (2.6529)  time: 0.3476  data: 0.0007  max mem: 17867
[19:49:46.509391] Epoch: [148] Total time: 0:29:10 (0.3497 s / it)
[19:49:46.513044] Averaged stats: lr: 0.000192  loss: 2.7203 (2.6535)
[19:49:47.552809] Test:  [   0/1563]  eta: 0:26:59  loss: 0.3920 (0.3920)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.0364  data: 0.8978  max mem: 17867
[19:50:52.575905] Test:  [ 500/1563]  eta: 0:02:20  loss: 1.0043 (0.7845)  acc1: 78.1250 (80.6325)  acc5: 96.8750 (96.1639)  time: 0.1300  data: 0.0002  max mem: 17867
[19:51:57.561303] Test:  [1000/1563]  eta: 0:01:13  loss: 1.3121 (0.9243)  acc1: 65.6250 (77.4257)  acc5: 93.7500 (94.3275)  time: 0.1299  data: 0.0002  max mem: 17867
[19:53:02.628032] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5503 (1.0023)  acc1: 84.3750 (75.5746)  acc5: 96.8750 (93.3482)  time: 0.1300  data: 0.0002  max mem: 17867
[19:53:10.609858] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4656 (1.0023)  acc1: 90.6250 (75.5980)  acc5: 100.0000 (93.3500)  time: 0.1262  data: 0.0001  max mem: 17867
[19:53:10.685048] Test: Total time: 0:03:24 (0.1306 s / it)
[19:53:10.856485] * Acc@1 75.598 Acc@5 93.350 loss 1.002
[19:53:10.856651] Accuracy of the network on the 50000 test images: 75.6%
[19:53:10.856673] Max accuracy: 75.60%
[19:53:10.903679] log_dir: ./output_dir_cml_spikformer
[19:53:12.581238] Epoch: [149]  [   0/5004]  eta: 2:19:49  lr: 0.000192  loss: 2.4543 (2.4543)  time: 1.6765  data: 0.9966  max mem: 17867
[20:04:51.879319] Epoch: [149]  [2000/5004]  eta: 0:17:32  lr: 0.000189  loss: 2.7591 (2.6413)  time: 0.3483  data: 0.0002  max mem: 17867
[20:16:31.939226] Epoch: [149]  [4000/5004]  eta: 0:05:51  lr: 0.000187  loss: 2.6502 (2.6430)  time: 0.3472  data: 0.0002  max mem: 17867
[20:22:22.803667] Epoch: [149]  [5003/5004]  eta: 0:00:00  lr: 0.000185  loss: 2.4488 (2.6453)  time: 0.3460  data: 0.0011  max mem: 17867
[20:22:23.168450] Epoch: [149] Total time: 0:29:12 (0.3502 s / it)
[20:22:23.170620] Averaged stats: lr: 0.000185  loss: 2.4488 (2.6445)
[20:22:24.334972] Test:  [   0/1563]  eta: 0:30:14  loss: 0.4584 (0.4584)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.1608  data: 1.0235  max mem: 17867
[20:23:29.589338] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.6242 (0.7866)  acc1: 78.1250 (80.4828)  acc5: 96.8750 (96.1639)  time: 0.1310  data: 0.0002  max mem: 17867
[20:24:34.610992] Test:  [1000/1563]  eta: 0:01:13  loss: 1.2307 (0.9183)  acc1: 68.7500 (77.6130)  acc5: 93.7500 (94.5211)  time: 0.1300  data: 0.0002  max mem: 17867
[20:25:39.588987] Test:  [1500/1563]  eta: 0:00:08  loss: 0.6238 (0.9982)  acc1: 87.5000 (75.8848)  acc5: 96.8750 (93.4689)  time: 0.1299  data: 0.0002  max mem: 17867
[20:25:47.580883] Test:  [1562/1563]  eta: 0:00:00  loss: 0.3890 (0.9996)  acc1: 90.6250 (75.8480)  acc5: 96.8750 (93.4580)  time: 0.1262  data: 0.0001  max mem: 17867
[20:25:47.652451] Test: Total time: 0:03:24 (0.1308 s / it)
[20:25:47.946298] * Acc@1 75.848 Acc@5 93.458 loss 1.000
[20:25:47.946443] Accuracy of the network on the 50000 test images: 75.8%
[20:25:47.946464] Max accuracy: 75.85%
[20:25:47.996118] log_dir: ./output_dir_cml_spikformer
[20:25:49.461989] Epoch: [150]  [   0/5004]  eta: 2:02:11  lr: 0.000185  loss: 2.8589 (2.8589)  time: 1.4651  data: 0.9278  max mem: 17867
[20:37:29.905520] Epoch: [150]  [2000/5004]  eta: 0:17:33  lr: 0.000182  loss: 2.5567 (2.6358)  time: 0.3518  data: 0.0002  max mem: 17867
[20:49:09.582471] Epoch: [150]  [4000/5004]  eta: 0:05:51  lr: 0.000180  loss: 2.5285 (2.6308)  time: 0.3485  data: 0.0002  max mem: 17867
[20:55:00.320791] Epoch: [150]  [5003/5004]  eta: 0:00:00  lr: 0.000178  loss: 2.6156 (2.6343)  time: 0.3531  data: 0.0007  max mem: 17867
[20:55:00.706940] Epoch: [150] Total time: 0:29:12 (0.3503 s / it)
[20:55:00.709321] Averaged stats: lr: 0.000178  loss: 2.6156 (2.6359)
[20:55:01.743527] Test:  [   0/1563]  eta: 0:26:50  loss: 0.4014 (0.4014)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.0307  data: 0.8793  max mem: 17867
[20:56:06.757867] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.8134 (0.8025)  acc1: 75.0000 (79.9838)  acc5: 96.8750 (95.9518)  time: 0.1303  data: 0.0002  max mem: 17867
[20:57:11.713799] Test:  [1000/1563]  eta: 0:01:13  loss: 0.9720 (0.9234)  acc1: 78.1250 (77.2665)  acc5: 96.8750 (94.3588)  time: 0.1304  data: 0.0002  max mem: 17867
[20:58:16.722098] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5190 (0.9964)  acc1: 87.5000 (75.6412)  acc5: 96.8750 (93.3836)  time: 0.1313  data: 0.0002  max mem: 17867
[20:58:24.703645] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4518 (0.9967)  acc1: 90.6250 (75.6500)  acc5: 96.8750 (93.3880)  time: 0.1261  data: 0.0001  max mem: 17867
[20:58:24.767083] Test: Total time: 0:03:24 (0.1306 s / it)
[20:58:25.197176] * Acc@1 75.650 Acc@5 93.388 loss 0.997
[20:58:25.197397] Accuracy of the network on the 50000 test images: 75.7%
[20:58:25.197418] Max accuracy: 75.85%
[20:58:25.243080] log_dir: ./output_dir_cml_spikformer
[20:58:26.772151] Epoch: [151]  [   0/5004]  eta: 2:07:27  lr: 0.000178  loss: 2.6452 (2.6452)  time: 1.5283  data: 1.1915  max mem: 17867
[21:10:07.554644] Epoch: [151]  [2000/5004]  eta: 0:17:34  lr: 0.000176  loss: 2.4729 (2.6301)  time: 0.3513  data: 0.0002  max mem: 17867
[21:21:48.130505] Epoch: [151]  [4000/5004]  eta: 0:05:52  lr: 0.000173  loss: 2.5708 (2.6243)  time: 0.3474  data: 0.0002  max mem: 17867
[21:27:39.544427] Epoch: [151]  [5003/5004]  eta: 0:00:00  lr: 0.000172  loss: 2.4285 (2.6249)  time: 0.3461  data: 0.0011  max mem: 17867
[21:27:39.939253] Epoch: [151] Total time: 0:29:14 (0.3507 s / it)
[21:27:39.940099] Averaged stats: lr: 0.000172  loss: 2.4285 (2.6292)
[21:27:41.464226] Test:  [   0/1563]  eta: 0:39:36  loss: 0.4424 (0.4424)  acc1: 90.6250 (90.6250)  acc5: 96.8750 (96.8750)  time: 1.5207  data: 1.3776  max mem: 17867
[21:28:46.480565] Test:  [ 500/1563]  eta: 0:02:21  loss: 0.7512 (0.7650)  acc1: 81.2500 (80.6075)  acc5: 96.8750 (96.2575)  time: 0.1301  data: 0.0002  max mem: 17867
[21:29:51.455083] Test:  [1000/1563]  eta: 0:01:13  loss: 1.0446 (0.8928)  acc1: 68.7500 (77.7597)  acc5: 93.7500 (94.4306)  time: 0.1299  data: 0.0002  max mem: 17867
[21:30:56.449324] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5439 (0.9651)  acc1: 87.5000 (76.0535)  acc5: 96.8750 (93.5043)  time: 0.1299  data: 0.0002  max mem: 17867
[21:31:04.438812] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4397 (0.9683)  acc1: 90.6250 (75.9860)  acc5: 96.8750 (93.4820)  time: 0.1262  data: 0.0001  max mem: 17867
[21:31:04.518261] Test: Total time: 0:03:24 (0.1309 s / it)
[21:31:04.842728] * Acc@1 75.986 Acc@5 93.482 loss 0.968
[21:31:04.842895] Accuracy of the network on the 50000 test images: 76.0%
[21:31:04.842918] Max accuracy: 75.99%
[21:31:04.859225] log_dir: ./output_dir_cml_spikformer
[21:31:06.335179] Epoch: [152]  [   0/5004]  eta: 2:02:59  lr: 0.000171  loss: 2.3605 (2.3605)  time: 1.4748  data: 1.0114  max mem: 17867
[21:42:47.179827] Epoch: [152]  [2000/5004]  eta: 0:17:34  lr: 0.000169  loss: 2.6053 (2.6172)  time: 0.3476  data: 0.0002  max mem: 17867
[21:54:27.259176] Epoch: [152]  [4000/5004]  eta: 0:05:51  lr: 0.000166  loss: 2.5341 (2.6170)  time: 0.3471  data: 0.0002  max mem: 17867
[22:00:18.229515] Epoch: [152]  [5003/5004]  eta: 0:00:00  lr: 0.000165  loss: 2.5133 (2.6177)  time: 0.3432  data: 0.0006  max mem: 17867
[22:00:18.617957] Epoch: [152] Total time: 0:29:13 (0.3505 s / it)
[22:00:18.618620] Averaged stats: lr: 0.000165  loss: 2.5133 (2.6202)
[22:00:19.684321] Test:  [   0/1563]  eta: 0:27:40  loss: 0.4104 (0.4104)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.0622  data: 0.9175  max mem: 17867
[22:01:24.689390] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.8147 (0.7929)  acc1: 81.2500 (80.1584)  acc5: 96.8750 (95.9019)  time: 0.1299  data: 0.0002  max mem: 17867
[22:02:29.661329] Test:  [1000/1563]  eta: 0:01:13  loss: 1.0774 (0.9194)  acc1: 71.8750 (77.2883)  acc5: 93.7500 (94.3307)  time: 0.1299  data: 0.0002  max mem: 17867
[22:03:34.628235] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5165 (0.9927)  acc1: 87.5000 (75.7224)  acc5: 96.8750 (93.3669)  time: 0.1298  data: 0.0002  max mem: 17867
[22:03:42.610764] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4198 (0.9937)  acc1: 90.6250 (75.6640)  acc5: 96.8750 (93.3720)  time: 0.1262  data: 0.0001  max mem: 17867
[22:03:42.671838] Test: Total time: 0:03:24 (0.1306 s / it)
[22:03:42.989261] * Acc@1 75.664 Acc@5 93.372 loss 0.994
[22:03:42.989436] Accuracy of the network on the 50000 test images: 75.7%
[22:03:42.989457] Max accuracy: 75.99%
[22:03:43.037206] log_dir: ./output_dir_cml_spikformer
[22:03:44.604143] Epoch: [153]  [   0/5004]  eta: 2:10:34  lr: 0.000165  loss: 2.5069 (2.5069)  time: 1.5656  data: 1.0016  max mem: 17867
[22:15:25.208408] Epoch: [153]  [2000/5004]  eta: 0:17:34  lr: 0.000162  loss: 2.4960 (2.6158)  time: 0.3524  data: 0.0002  max mem: 17867
[22:27:04.611256] Epoch: [153]  [4000/5004]  eta: 0:05:51  lr: 0.000160  loss: 2.6290 (2.6204)  time: 0.3473  data: 0.0002  max mem: 17867
[22:32:55.846977] Epoch: [153]  [5003/5004]  eta: 0:00:00  lr: 0.000158  loss: 2.5644 (2.6237)  time: 0.3477  data: 0.0011  max mem: 17867
[22:32:56.238824] Epoch: [153] Total time: 0:29:13 (0.3504 s / it)
[22:32:56.239537] Averaged stats: lr: 0.000158  loss: 2.5644 (2.6128)
[22:32:57.541635] Test:  [   0/1563]  eta: 0:33:48  loss: 0.5245 (0.5245)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.2981  data: 1.1609  max mem: 17867
[22:34:02.693593] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.7186 (0.7573)  acc1: 81.2500 (80.9132)  acc5: 96.8750 (96.3136)  time: 0.1304  data: 0.0003  max mem: 17867
[22:35:07.815895] Test:  [1000/1563]  eta: 0:01:13  loss: 1.0028 (0.8908)  acc1: 75.0000 (77.9751)  acc5: 93.7500 (94.5617)  time: 0.1303  data: 0.0002  max mem: 17867
[22:36:12.886260] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5538 (0.9714)  acc1: 84.3750 (76.1055)  acc5: 96.8750 (93.5314)  time: 0.1299  data: 0.0002  max mem: 17867
[22:36:20.883763] Test:  [1562/1563]  eta: 0:00:00  loss: 0.3951 (0.9723)  acc1: 93.7500 (76.0720)  acc5: 96.8750 (93.5300)  time: 0.1263  data: 0.0001  max mem: 17867
[22:36:20.964232] Test: Total time: 0:03:24 (0.1310 s / it)
[22:36:21.245161] * Acc@1 76.072 Acc@5 93.530 loss 0.972
[22:36:21.245309] Accuracy of the network on the 50000 test images: 76.1%
[22:36:21.245331] Max accuracy: 76.07%
[22:36:21.286405] log_dir: ./output_dir_cml_spikformer
[22:36:22.780978] Epoch: [154]  [   0/5004]  eta: 2:04:32  lr: 0.000158  loss: 2.1467 (2.1467)  time: 1.4932  data: 1.1154  max mem: 17867
[22:48:02.106860] Epoch: [154]  [2000/5004]  eta: 0:17:32  lr: 0.000156  loss: 2.5768 (2.6030)  time: 0.3472  data: 0.0002  max mem: 17867
[22:59:40.724610] Epoch: [154]  [4000/5004]  eta: 0:05:51  lr: 0.000153  loss: 2.5886 (2.6003)  time: 0.3513  data: 0.0002  max mem: 17867
[23:05:31.414160] Epoch: [154]  [5003/5004]  eta: 0:00:00  lr: 0.000152  loss: 2.6696 (2.6031)  time: 0.3484  data: 0.0006  max mem: 17867
[23:05:31.792274] Epoch: [154] Total time: 0:29:10 (0.3498 s / it)
[23:05:31.794526] Averaged stats: lr: 0.000152  loss: 2.6696 (2.6077)
[23:05:32.770941] Test:  [   0/1563]  eta: 0:25:20  loss: 0.4695 (0.4695)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 0.9729  data: 0.8328  max mem: 17867
[23:06:37.815012] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.7165 (0.7503)  acc1: 81.2500 (80.6387)  acc5: 96.8750 (96.2201)  time: 0.1299  data: 0.0002  max mem: 17867
[23:07:42.841697] Test:  [1000/1563]  eta: 0:01:13  loss: 1.1653 (0.8870)  acc1: 65.6250 (77.8097)  acc5: 93.7500 (94.5180)  time: 0.1299  data: 0.0002  max mem: 17867
[23:08:47.839628] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5897 (0.9621)  acc1: 84.3750 (76.0826)  acc5: 100.0000 (93.5168)  time: 0.1299  data: 0.0002  max mem: 17867
[23:08:55.825002] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4079 (0.9641)  acc1: 90.6250 (76.0180)  acc5: 100.0000 (93.5340)  time: 0.1262  data: 0.0001  max mem: 17867
[23:08:55.893137] Test: Total time: 0:03:24 (0.1306 s / it)
[23:08:56.228634] * Acc@1 76.018 Acc@5 93.534 loss 0.964
[23:08:56.228766] Accuracy of the network on the 50000 test images: 76.0%
[23:08:56.228787] Max accuracy: 76.07%
[23:08:56.278453] log_dir: ./output_dir_cml_spikformer
[23:08:57.781595] Epoch: [155]  [   0/5004]  eta: 2:05:17  lr: 0.000152  loss: 2.3990 (2.3990)  time: 1.5023  data: 1.1556  max mem: 17867
[23:20:37.847545] Epoch: [155]  [2000/5004]  eta: 0:17:33  lr: 0.000149  loss: 2.5498 (2.5846)  time: 0.3547  data: 0.0002  max mem: 17867
[23:32:17.077699] Epoch: [155]  [4000/5004]  eta: 0:05:51  lr: 0.000147  loss: 2.6812 (2.5942)  time: 0.3502  data: 0.0002  max mem: 17867
[23:38:07.847269] Epoch: [155]  [5003/5004]  eta: 0:00:00  lr: 0.000145  loss: 2.5285 (2.5965)  time: 0.3464  data: 0.0011  max mem: 17867
[23:38:08.216341] Epoch: [155] Total time: 0:29:11 (0.3501 s / it)
[23:38:08.224318] Averaged stats: lr: 0.000145  loss: 2.5285 (2.5950)
[23:38:09.270047] Test:  [   0/1563]  eta: 0:27:09  loss: 0.2874 (0.2874)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.0423  data: 0.8988  max mem: 17867
[23:39:14.368066] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.7556 (0.7678)  acc1: 81.2500 (80.7884)  acc5: 96.8750 (96.3323)  time: 0.1304  data: 0.0002  max mem: 17867
[23:40:19.393071] Test:  [1000/1563]  eta: 0:01:13  loss: 1.1087 (0.8942)  acc1: 65.6250 (77.8409)  acc5: 93.7500 (94.6116)  time: 0.1299  data: 0.0002  max mem: 17867
[23:41:24.431532] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5706 (0.9679)  acc1: 87.5000 (76.2221)  acc5: 93.7500 (93.6542)  time: 0.1300  data: 0.0002  max mem: 17867
[23:41:32.418998] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4236 (0.9684)  acc1: 90.6250 (76.2320)  acc5: 96.8750 (93.6660)  time: 0.1262  data: 0.0001  max mem: 17867
[23:41:32.491620] Test: Total time: 0:03:24 (0.1307 s / it)
[23:41:32.836163] * Acc@1 76.232 Acc@5 93.666 loss 0.968
[23:41:32.836311] Accuracy of the network on the 50000 test images: 76.2%
[23:41:32.836339] Max accuracy: 76.23%
[23:41:32.877806] log_dir: ./output_dir_cml_spikformer
[23:41:34.309161] Epoch: [156]  [   0/5004]  eta: 1:59:17  lr: 0.000145  loss: 2.7333 (2.7333)  time: 1.4304  data: 1.0372  max mem: 17867
[23:53:14.049347] Epoch: [156]  [2000/5004]  eta: 0:17:32  lr: 0.000143  loss: 2.5941 (2.5986)  time: 0.3459  data: 0.0002  max mem: 17867
[00:04:53.220142] Epoch: [156]  [4000/5004]  eta: 0:05:51  lr: 0.000140  loss: 2.5432 (2.5976)  time: 0.3473  data: 0.0002  max mem: 17867
[00:10:43.585987] Epoch: [156]  [5003/5004]  eta: 0:00:00  lr: 0.000139  loss: 2.5449 (2.5950)  time: 0.3443  data: 0.0011  max mem: 17867
[00:10:43.950903] Epoch: [156] Total time: 0:29:11 (0.3499 s / it)
[00:10:43.957732] Averaged stats: lr: 0.000139  loss: 2.5449 (2.5889)
[00:10:45.009234] Test:  [   0/1563]  eta: 0:27:17  loss: 0.3008 (0.3008)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.0479  data: 0.8941  max mem: 17867
[00:11:50.065265] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.7063 (0.7563)  acc1: 81.2500 (80.7822)  acc5: 96.8750 (96.4072)  time: 0.1300  data: 0.0002  max mem: 17867
[00:12:55.066808] Test:  [1000/1563]  eta: 0:01:13  loss: 1.1383 (0.8791)  acc1: 75.0000 (78.1625)  acc5: 93.7500 (94.7147)  time: 0.1302  data: 0.0002  max mem: 17867
[00:14:00.056686] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5363 (0.9548)  acc1: 87.5000 (76.4907)  acc5: 96.8750 (93.7771)  time: 0.1300  data: 0.0002  max mem: 17867
[00:14:08.052290] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4344 (0.9574)  acc1: 90.6250 (76.4300)  acc5: 100.0000 (93.7680)  time: 0.1262  data: 0.0001  max mem: 17867
[00:14:08.137974] Test: Total time: 0:03:24 (0.1306 s / it)
[00:14:08.284568] * Acc@1 76.430 Acc@5 93.768 loss 0.957
[00:14:08.284736] Accuracy of the network on the 50000 test images: 76.4%
[00:14:08.284757] Max accuracy: 76.43%
[00:14:08.330368] log_dir: ./output_dir_cml_spikformer
[00:14:09.852489] Epoch: [157]  [   0/5004]  eta: 2:06:53  lr: 0.000139  loss: 2.4357 (2.4357)  time: 1.5215  data: 1.0620  max mem: 17867
[00:25:50.472762] Epoch: [157]  [2000/5004]  eta: 0:17:34  lr: 0.000137  loss: 2.5308 (2.5780)  time: 0.3525  data: 0.0002  max mem: 17867
[00:37:30.927504] Epoch: [157]  [4000/5004]  eta: 0:05:51  lr: 0.000134  loss: 2.4851 (2.5793)  time: 0.3515  data: 0.0006  max mem: 17867
[00:43:22.010762] Epoch: [157]  [5003/5004]  eta: 0:00:00  lr: 0.000133  loss: 2.5596 (2.5817)  time: 0.3451  data: 0.0011  max mem: 17867
[00:43:22.395075] Epoch: [157] Total time: 0:29:14 (0.3505 s / it)
[00:43:22.395734] Averaged stats: lr: 0.000133  loss: 2.5596 (2.5791)
[00:43:23.394512] Test:  [   0/1563]  eta: 0:25:55  loss: 0.3729 (0.3729)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 0.9952  data: 0.8354  max mem: 17867
[00:44:28.431842] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.8644 (0.7496)  acc1: 78.1250 (81.2999)  acc5: 96.8750 (96.2824)  time: 0.1300  data: 0.0002  max mem: 17867
[00:45:33.427314] Test:  [1000/1563]  eta: 0:01:13  loss: 1.1381 (0.8799)  acc1: 71.8750 (78.4434)  acc5: 93.7500 (94.5960)  time: 0.1299  data: 0.0002  max mem: 17867
[00:46:38.456564] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5342 (0.9565)  acc1: 90.6250 (76.6926)  acc5: 96.8750 (93.6896)  time: 0.1301  data: 0.0002  max mem: 17867
[00:46:46.436673] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4018 (0.9593)  acc1: 90.6250 (76.6160)  acc5: 96.8750 (93.6880)  time: 0.1261  data: 0.0001  max mem: 17867
[00:46:46.514452] Test: Total time: 0:03:24 (0.1306 s / it)
[00:46:46.750611] * Acc@1 76.616 Acc@5 93.688 loss 0.959
[00:46:46.750744] Accuracy of the network on the 50000 test images: 76.6%
[00:46:46.750765] Max accuracy: 76.62%
[00:46:46.777150] log_dir: ./output_dir_cml_spikformer
[00:46:48.660752] Epoch: [158]  [   0/5004]  eta: 2:37:01  lr: 0.000133  loss: 2.4159 (2.4159)  time: 1.8828  data: 1.5475  max mem: 17867
[00:58:30.525133] Epoch: [158]  [2000/5004]  eta: 0:17:36  lr: 0.000131  loss: 2.4777 (2.5595)  time: 0.3534  data: 0.0004  max mem: 17867
[01:10:11.191845] Epoch: [158]  [4000/5004]  eta: 0:05:52  lr: 0.000128  loss: 2.5007 (2.5688)  time: 0.3480  data: 0.0002  max mem: 17867
[01:16:02.347121] Epoch: [158]  [5003/5004]  eta: 0:00:00  lr: 0.000127  loss: 2.5172 (2.5687)  time: 0.3469  data: 0.0011  max mem: 17867
[01:16:02.732773] Epoch: [158] Total time: 0:29:15 (0.3509 s / it)
[01:16:02.770446] Averaged stats: lr: 0.000127  loss: 2.5172 (2.5713)
[01:16:04.079737] Test:  [   0/1563]  eta: 0:34:00  loss: 0.4589 (0.4589)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.3055  data: 1.0491  max mem: 17867
[01:17:09.129384] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.6540 (0.7575)  acc1: 81.2500 (81.0878)  acc5: 96.8750 (96.3386)  time: 0.1299  data: 0.0002  max mem: 17867
[01:18:14.113900] Test:  [1000/1563]  eta: 0:01:13  loss: 1.1201 (0.8876)  acc1: 71.8750 (78.2530)  acc5: 93.7500 (94.5430)  time: 0.1298  data: 0.0002  max mem: 17867
[01:19:19.199245] Test:  [1500/1563]  eta: 0:00:08  loss: 0.4979 (0.9635)  acc1: 87.5000 (76.4470)  acc5: 100.0000 (93.6521)  time: 0.1298  data: 0.0002  max mem: 17867
[01:19:27.301061] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4869 (0.9657)  acc1: 93.7500 (76.3340)  acc5: 100.0000 (93.6560)  time: 0.1323  data: 0.0001  max mem: 17867
[01:19:27.376165] Test: Total time: 0:03:24 (0.1309 s / it)
[01:19:27.658504] * Acc@1 76.334 Acc@5 93.656 loss 0.966
[01:19:27.658641] Accuracy of the network on the 50000 test images: 76.3%
[01:19:27.658660] Max accuracy: 76.62%
[01:19:27.666607] log_dir: ./output_dir_cml_spikformer
[01:19:29.343140] Epoch: [159]  [   0/5004]  eta: 2:19:46  lr: 0.000127  loss: 2.6729 (2.6729)  time: 1.6759  data: 1.0171  max mem: 17867
[01:31:10.002291] Epoch: [159]  [2000/5004]  eta: 0:17:34  lr: 0.000125  loss: 2.6873 (2.5665)  time: 0.3507  data: 0.0002  max mem: 17867
[01:42:49.868696] Epoch: [159]  [4000/5004]  eta: 0:05:51  lr: 0.000122  loss: 2.5725 (2.5695)  time: 0.3467  data: 0.0002  max mem: 17867
[01:48:41.210973] Epoch: [159]  [5003/5004]  eta: 0:00:00  lr: 0.000121  loss: 2.6894 (2.5733)  time: 0.3464  data: 0.0006  max mem: 17867
[01:48:41.596106] Epoch: [159] Total time: 0:29:13 (0.3505 s / it)
[01:48:41.613015] Averaged stats: lr: 0.000121  loss: 2.6894 (2.5658)
[01:48:42.721500] Test:  [   0/1563]  eta: 0:28:45  loss: 0.3055 (0.3055)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.1040  data: 0.9236  max mem: 17867
[01:49:47.747790] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.7238 (0.7460)  acc1: 75.0000 (81.2188)  acc5: 96.8750 (96.4197)  time: 0.1299  data: 0.0002  max mem: 17867
[01:50:52.732618] Test:  [1000/1563]  eta: 0:01:13  loss: 1.2992 (0.8746)  acc1: 65.6250 (78.1593)  acc5: 90.6250 (94.7115)  time: 0.1299  data: 0.0002  max mem: 17867
[01:51:57.723508] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5597 (0.9492)  acc1: 84.3750 (76.5698)  acc5: 96.8750 (93.7292)  time: 0.1299  data: 0.0002  max mem: 17867
[01:52:05.711156] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4386 (0.9516)  acc1: 90.6250 (76.4880)  acc5: 96.8750 (93.7060)  time: 0.1262  data: 0.0001  max mem: 17867
[01:52:05.771404] Test: Total time: 0:03:24 (0.1306 s / it)
[01:52:06.032807] * Acc@1 76.488 Acc@5 93.706 loss 0.952
[01:52:06.032975] Accuracy of the network on the 50000 test images: 76.5%
[01:52:06.033000] Max accuracy: 76.62%
[01:52:06.071651] log_dir: ./output_dir_cml_spikformer
[01:52:07.606114] Epoch: [160]  [   0/5004]  eta: 2:07:54  lr: 0.000121  loss: 2.2998 (2.2998)  time: 1.5337  data: 1.0726  max mem: 17867
[02:03:48.631801] Epoch: [160]  [2000/5004]  eta: 0:17:34  lr: 0.000119  loss: 2.5384 (2.5409)  time: 0.3473  data: 0.0002  max mem: 17867
[02:15:28.946561] Epoch: [160]  [4000/5004]  eta: 0:05:51  lr: 0.000117  loss: 2.6386 (2.5494)  time: 0.3483  data: 0.0002  max mem: 17867
[02:21:19.756519] Epoch: [160]  [5003/5004]  eta: 0:00:00  lr: 0.000115  loss: 2.6040 (2.5543)  time: 0.3458  data: 0.0011  max mem: 17867
[02:21:20.136501] Epoch: [160] Total time: 0:29:14 (0.3505 s / it)
[02:21:20.143939] Averaged stats: lr: 0.000115  loss: 2.6040 (2.5573)
[02:21:21.414870] Test:  [   0/1563]  eta: 0:33:00  loss: 0.3191 (0.3191)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.2673  data: 1.1243  max mem: 17867
[02:22:26.528539] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.8079 (0.7493)  acc1: 81.2500 (81.4621)  acc5: 96.8750 (96.4384)  time: 0.1299  data: 0.0002  max mem: 17867
[02:23:31.530073] Test:  [1000/1563]  eta: 0:01:13  loss: 1.1125 (0.8793)  acc1: 75.0000 (78.4434)  acc5: 93.7500 (94.7802)  time: 0.1299  data: 0.0002  max mem: 17867
[02:24:36.528216] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5981 (0.9526)  acc1: 87.5000 (76.7509)  acc5: 96.8750 (93.8395)  time: 0.1300  data: 0.0002  max mem: 17867
[02:24:44.511130] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4698 (0.9533)  acc1: 90.6250 (76.7000)  acc5: 96.8750 (93.8500)  time: 0.1262  data: 0.0001  max mem: 17867
[02:24:44.597252] Test: Total time: 0:03:24 (0.1308 s / it)
[02:24:44.685191] * Acc@1 76.700 Acc@5 93.850 loss 0.953
[02:24:44.685332] Accuracy of the network on the 50000 test images: 76.7%
[02:24:44.685354] Max accuracy: 76.70%
[02:24:44.709851] log_dir: ./output_dir_cml_spikformer
[02:24:46.221331] Epoch: [161]  [   0/5004]  eta: 2:05:57  lr: 0.000115  loss: 2.4669 (2.4669)  time: 1.5104  data: 0.9559  max mem: 17867
[02:36:27.359887] Epoch: [161]  [2000/5004]  eta: 0:17:34  lr: 0.000113  loss: 2.4660 (2.5494)  time: 0.3523  data: 0.0002  max mem: 17867
[02:48:07.782354] Epoch: [161]  [4000/5004]  eta: 0:05:52  lr: 0.000111  loss: 2.4951 (2.5475)  time: 0.3488  data: 0.0002  max mem: 17867
[02:53:58.454522] Epoch: [161]  [5003/5004]  eta: 0:00:00  lr: 0.000110  loss: 2.4839 (2.5448)  time: 0.3462  data: 0.0011  max mem: 17867
[02:53:58.886979] Epoch: [161] Total time: 0:29:14 (0.3506 s / it)
[02:53:58.890348] Averaged stats: lr: 0.000110  loss: 2.4839 (2.5494)
[02:53:59.917771] Test:  [   0/1563]  eta: 0:26:40  loss: 0.4291 (0.4291)  acc1: 87.5000 (87.5000)  acc5: 96.8750 (96.8750)  time: 1.0239  data: 0.8865  max mem: 17867
[02:55:04.940705] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.9178 (0.7355)  acc1: 78.1250 (81.5744)  acc5: 96.8750 (96.2887)  time: 0.1299  data: 0.0002  max mem: 17867
[02:56:09.906759] Test:  [1000/1563]  eta: 0:01:13  loss: 1.0683 (0.8537)  acc1: 71.8750 (78.7744)  acc5: 90.6250 (94.8021)  time: 0.1298  data: 0.0002  max mem: 17867
[02:57:14.890967] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5199 (0.9336)  acc1: 84.3750 (76.9799)  acc5: 96.8750 (93.7771)  time: 0.1299  data: 0.0002  max mem: 17867
[02:57:22.868305] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4238 (0.9351)  acc1: 90.6250 (76.9100)  acc5: 96.8750 (93.7800)  time: 0.1261  data: 0.0001  max mem: 17867
[02:57:22.937027] Test: Total time: 0:03:24 (0.1305 s / it)
[02:57:23.220458] * Acc@1 76.910 Acc@5 93.780 loss 0.935
[02:57:23.220599] Accuracy of the network on the 50000 test images: 76.9%
[02:57:23.220620] Max accuracy: 76.91%
[02:57:23.265609] log_dir: ./output_dir_cml_spikformer
[02:57:24.733002] Epoch: [162]  [   0/5004]  eta: 2:02:18  lr: 0.000110  loss: 2.3190 (2.3190)  time: 1.4665  data: 1.0508  max mem: 17867
[03:09:03.681315] Epoch: [162]  [2000/5004]  eta: 0:17:31  lr: 0.000108  loss: 2.6906 (2.5506)  time: 0.3519  data: 0.0002  max mem: 17867
[03:20:43.070190] Epoch: [162]  [4000/5004]  eta: 0:05:51  lr: 0.000105  loss: 2.5804 (2.5404)  time: 0.3461  data: 0.0002  max mem: 17867
[03:26:33.702924] Epoch: [162]  [5003/5004]  eta: 0:00:00  lr: 0.000104  loss: 2.5668 (2.5436)  time: 0.3451  data: 0.0006  max mem: 17867
[03:26:34.056799] Epoch: [162] Total time: 0:29:10 (0.3499 s / it)
[03:26:34.062262] Averaged stats: lr: 0.000104  loss: 2.5668 (2.5399)
[03:26:35.105271] Test:  [   0/1563]  eta: 0:27:04  loss: 0.3729 (0.3729)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.0392  data: 0.8676  max mem: 17867
[03:27:40.185514] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.7804 (0.7608)  acc1: 81.2500 (81.3373)  acc5: 96.8750 (96.3573)  time: 0.1299  data: 0.0002  max mem: 17867
[03:28:45.185217] Test:  [1000/1563]  eta: 0:01:13  loss: 1.1952 (0.8849)  acc1: 68.7500 (78.5184)  acc5: 93.7500 (94.7334)  time: 0.1300  data: 0.0002  max mem: 17867
[03:29:50.200271] Test:  [1500/1563]  eta: 0:00:08  loss: 0.4582 (0.9616)  acc1: 87.5000 (76.6885)  acc5: 96.8750 (93.8020)  time: 0.1300  data: 0.0002  max mem: 17867
[03:29:58.183776] Test:  [1562/1563]  eta: 0:00:00  loss: 0.3955 (0.9623)  acc1: 90.6250 (76.6620)  acc5: 100.0000 (93.8160)  time: 0.1262  data: 0.0001  max mem: 17867
[03:29:58.256954] Test: Total time: 0:03:24 (0.1306 s / it)
[03:29:58.349494] * Acc@1 76.662 Acc@5 93.816 loss 0.962
[03:29:58.349646] Accuracy of the network on the 50000 test images: 76.7%
[03:29:58.349671] Max accuracy: 76.91%
[03:29:58.375857] log_dir: ./output_dir_cml_spikformer
[03:30:00.061327] Epoch: [163]  [   0/5004]  eta: 2:20:29  lr: 0.000104  loss: 1.9935 (1.9935)  time: 1.6846  data: 1.1454  max mem: 17867
[03:41:39.840522] Epoch: [163]  [2000/5004]  eta: 0:17:33  lr: 0.000102  loss: 2.5697 (2.5360)  time: 0.3467  data: 0.0002  max mem: 17867
[03:53:19.547031] Epoch: [163]  [4000/5004]  eta: 0:05:51  lr: 0.000100  loss: 2.5621 (2.5261)  time: 0.3452  data: 0.0002  max mem: 17867
[03:59:10.401474] Epoch: [163]  [5003/5004]  eta: 0:00:00  lr: 0.000099  loss: 2.4537 (2.5305)  time: 0.3431  data: 0.0011  max mem: 17867
[03:59:10.797214] Epoch: [163] Total time: 0:29:12 (0.3502 s / it)
[03:59:10.804841] Averaged stats: lr: 0.000099  loss: 2.4537 (2.5353)
[03:59:12.270313] Test:  [   0/1563]  eta: 0:38:05  loss: 0.4301 (0.4301)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.4620  data: 1.3243  max mem: 17867
[04:00:17.298123] Test:  [ 500/1563]  eta: 0:02:21  loss: 0.7971 (0.7325)  acc1: 78.1250 (81.7615)  acc5: 96.8750 (96.4758)  time: 0.1302  data: 0.0002  max mem: 17867
[04:01:22.315951] Test:  [1000/1563]  eta: 0:01:13  loss: 1.2669 (0.8624)  acc1: 65.6250 (78.8274)  acc5: 93.7500 (94.7709)  time: 0.1300  data: 0.0002  max mem: 17867
[04:02:27.427918] Test:  [1500/1563]  eta: 0:00:08  loss: 0.4397 (0.9320)  acc1: 87.5000 (77.1153)  acc5: 96.8750 (93.9082)  time: 0.1301  data: 0.0002  max mem: 17867
[04:02:35.415915] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4684 (0.9330)  acc1: 90.6250 (77.0520)  acc5: 100.0000 (93.9300)  time: 0.1265  data: 0.0001  max mem: 17867
[04:02:35.532804] Test: Total time: 0:03:24 (0.1310 s / it)
[04:02:35.648073] * Acc@1 77.052 Acc@5 93.930 loss 0.933
[04:02:35.648209] Accuracy of the network on the 50000 test images: 77.1%
[04:02:35.648229] Max accuracy: 77.05%
[04:02:35.687225] log_dir: ./output_dir_cml_spikformer
[04:02:37.204820] Epoch: [164]  [   0/5004]  eta: 2:06:29  lr: 0.000099  loss: 2.2903 (2.2903)  time: 1.5167  data: 0.9231  max mem: 17867
[04:14:17.160274] Epoch: [164]  [2000/5004]  eta: 0:17:33  lr: 0.000097  loss: 2.5411 (2.5288)  time: 0.3498  data: 0.0002  max mem: 17867
[04:25:57.219582] Epoch: [164]  [4000/5004]  eta: 0:05:51  lr: 0.000095  loss: 2.5079 (2.5276)  time: 0.3493  data: 0.0002  max mem: 17867
[04:31:47.912090] Epoch: [164]  [5003/5004]  eta: 0:00:00  lr: 0.000094  loss: 2.4410 (2.5258)  time: 0.3442  data: 0.0011  max mem: 17867
[04:31:48.312784] Epoch: [164] Total time: 0:29:12 (0.3502 s / it)
[04:31:48.324196] Averaged stats: lr: 0.000094  loss: 2.4410 (2.5255)
[04:31:49.330716] Test:  [   0/1563]  eta: 0:26:07  loss: 0.4272 (0.4272)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.0030  data: 0.8467  max mem: 17867
[04:32:54.396804] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.8090 (0.7370)  acc1: 78.1250 (81.5494)  acc5: 96.8750 (96.5382)  time: 0.1309  data: 0.0002  max mem: 17867
[04:33:59.441618] Test:  [1000/1563]  eta: 0:01:13  loss: 1.1023 (0.8574)  acc1: 68.7500 (78.8399)  acc5: 90.6250 (94.9332)  time: 0.1299  data: 0.0002  max mem: 17867
[04:35:04.430863] Test:  [1500/1563]  eta: 0:00:08  loss: 0.4776 (0.9300)  acc1: 84.3750 (77.2027)  acc5: 96.8750 (94.0123)  time: 0.1299  data: 0.0002  max mem: 17867
[04:35:12.424852] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4529 (0.9322)  acc1: 90.6250 (77.1240)  acc5: 100.0000 (93.9820)  time: 0.1266  data: 0.0001  max mem: 17867
[04:35:12.520062] Test: Total time: 0:03:24 (0.1306 s / it)
[04:35:12.696197] * Acc@1 77.124 Acc@5 93.982 loss 0.932
[04:35:12.696418] Accuracy of the network on the 50000 test images: 77.1%
[04:35:12.696442] Max accuracy: 77.12%
[04:35:12.722693] log_dir: ./output_dir_cml_spikformer
[04:35:14.244895] Epoch: [165]  [   0/5004]  eta: 2:06:52  lr: 0.000094  loss: 2.7075 (2.7075)  time: 1.5213  data: 1.0398  max mem: 17867
[04:46:54.802795] Epoch: [165]  [2000/5004]  eta: 0:17:33  lr: 0.000092  loss: 2.6006 (2.5056)  time: 0.3542  data: 0.0002  max mem: 17867
[04:58:35.354926] Epoch: [165]  [4000/5004]  eta: 0:05:51  lr: 0.000090  loss: 2.4450 (2.5083)  time: 0.3494  data: 0.0003  max mem: 17867
[05:04:26.392016] Epoch: [165]  [5003/5004]  eta: 0:00:00  lr: 0.000089  loss: 2.6456 (2.5153)  time: 0.3481  data: 0.0006  max mem: 17867
[05:04:26.782062] Epoch: [165] Total time: 0:29:14 (0.3505 s / it)
[05:04:26.789932] Averaged stats: lr: 0.000089  loss: 2.6456 (2.5183)
[05:04:27.847753] Test:  [   0/1563]  eta: 0:27:27  loss: 0.3986 (0.3986)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.0541  data: 0.9181  max mem: 17867
[05:05:32.870772] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.8238 (0.7337)  acc1: 78.1250 (81.7552)  acc5: 96.8750 (96.5257)  time: 0.1300  data: 0.0002  max mem: 17867
[05:06:37.872648] Test:  [1000/1563]  eta: 0:01:13  loss: 0.9839 (0.8610)  acc1: 68.7500 (78.8836)  acc5: 93.7500 (94.8926)  time: 0.1299  data: 0.0002  max mem: 17867
[05:07:42.851936] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5347 (0.9346)  acc1: 84.3750 (77.2797)  acc5: 100.0000 (93.9582)  time: 0.1299  data: 0.0002  max mem: 17867
[05:07:50.902802] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4351 (0.9375)  acc1: 90.6250 (77.1860)  acc5: 100.0000 (93.9460)  time: 0.1296  data: 0.0001  max mem: 17867
[05:07:50.972630] Test: Total time: 0:03:24 (0.1306 s / it)
[05:07:51.212559] * Acc@1 77.186 Acc@5 93.946 loss 0.937
[05:07:51.212733] Accuracy of the network on the 50000 test images: 77.2%
[05:07:51.212756] Max accuracy: 77.19%
[05:07:51.246848] log_dir: ./output_dir_cml_spikformer
[05:07:52.731617] Epoch: [166]  [   0/5004]  eta: 2:03:45  lr: 0.000089  loss: 2.0816 (2.0816)  time: 1.4839  data: 1.0391  max mem: 17867
[05:19:34.995734] Epoch: [166]  [2000/5004]  eta: 0:17:36  lr: 0.000087  loss: 2.5343 (2.5105)  time: 0.3498  data: 0.0002  max mem: 17867
[05:31:15.692823] Epoch: [166]  [4000/5004]  eta: 0:05:52  lr: 0.000085  loss: 2.3802 (2.5034)  time: 0.3446  data: 0.0002  max mem: 17867
[05:37:07.087757] Epoch: [166]  [5003/5004]  eta: 0:00:00  lr: 0.000084  loss: 2.3786 (2.5057)  time: 0.3457  data: 0.0006  max mem: 17867
[05:37:07.485168] Epoch: [166] Total time: 0:29:16 (0.3510 s / it)
[05:37:07.492154] Averaged stats: lr: 0.000084  loss: 2.3786 (2.5097)
[05:37:08.593734] Test:  [   0/1563]  eta: 0:28:36  loss: 0.2916 (0.2916)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.0979  data: 0.9518  max mem: 17867
[05:38:13.580348] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.8972 (0.7309)  acc1: 78.1250 (82.0172)  acc5: 96.8750 (96.6629)  time: 0.1299  data: 0.0002  max mem: 17867
[05:39:18.554664] Test:  [1000/1563]  eta: 0:01:13  loss: 1.0642 (0.8604)  acc1: 68.7500 (78.9866)  acc5: 93.7500 (94.8895)  time: 0.1299  data: 0.0002  max mem: 17867
[05:40:23.602219] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5025 (0.9327)  acc1: 87.5000 (77.2672)  acc5: 96.8750 (93.9665)  time: 0.1298  data: 0.0002  max mem: 17867
[05:40:31.580035] Test:  [1562/1563]  eta: 0:00:00  loss: 0.3769 (0.9340)  acc1: 90.6250 (77.2060)  acc5: 96.8750 (93.9660)  time: 0.1261  data: 0.0001  max mem: 17867
[05:40:31.669351] Test: Total time: 0:03:24 (0.1306 s / it)
[05:40:31.852510] * Acc@1 77.206 Acc@5 93.966 loss 0.934
[05:40:31.852649] Accuracy of the network on the 50000 test images: 77.2%
[05:40:31.852670] Max accuracy: 77.21%
[05:40:31.860181] log_dir: ./output_dir_cml_spikformer
[05:40:33.414914] Epoch: [167]  [   0/5004]  eta: 2:09:35  lr: 0.000084  loss: 2.0849 (2.0849)  time: 1.5538  data: 1.0896  max mem: 17867
[05:52:14.561237] Epoch: [167]  [2000/5004]  eta: 0:17:34  lr: 0.000082  loss: 2.4063 (2.4968)  time: 0.3526  data: 0.0002  max mem: 17867
[06:03:55.134632] Epoch: [167]  [4000/5004]  eta: 0:05:52  lr: 0.000080  loss: 2.3285 (2.4977)  time: 0.3572  data: 0.0002  max mem: 17867
[06:09:46.213039] Epoch: [167]  [5003/5004]  eta: 0:00:00  lr: 0.000079  loss: 2.6288 (2.4984)  time: 0.3438  data: 0.0006  max mem: 17867
[06:09:46.579222] Epoch: [167] Total time: 0:29:14 (0.3507 s / it)
[06:09:46.580037] Averaged stats: lr: 0.000079  loss: 2.6288 (2.5014)
[06:09:47.600862] Test:  [   0/1563]  eta: 0:26:29  loss: 0.4245 (0.4245)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.0173  data: 0.8800  max mem: 17867
[06:10:52.690598] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.7792 (0.7375)  acc1: 78.1250 (81.8239)  acc5: 96.8750 (96.5444)  time: 0.1302  data: 0.0002  max mem: 17867
[06:11:57.715530] Test:  [1000/1563]  eta: 0:01:13  loss: 1.0176 (0.8579)  acc1: 75.0000 (79.0959)  acc5: 93.7500 (94.8676)  time: 0.1303  data: 0.0002  max mem: 17867
[06:13:02.737099] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5564 (0.9292)  acc1: 87.5000 (77.4421)  acc5: 96.8750 (93.9894)  time: 0.1304  data: 0.0003  max mem: 17867
[06:13:10.721246] Test:  [1562/1563]  eta: 0:00:00  loss: 0.3710 (0.9307)  acc1: 90.6250 (77.3860)  acc5: 100.0000 (93.9900)  time: 0.1262  data: 0.0001  max mem: 17867
[06:13:10.789579] Test: Total time: 0:03:24 (0.1307 s / it)
[06:13:10.991132] * Acc@1 77.386 Acc@5 93.990 loss 0.931
[06:13:10.991269] Accuracy of the network on the 50000 test images: 77.4%
[06:13:10.991290] Max accuracy: 77.39%
[06:13:11.026762] log_dir: ./output_dir_cml_spikformer
[06:13:12.589033] Epoch: [168]  [   0/5004]  eta: 2:10:12  lr: 0.000079  loss: 2.4380 (2.4380)  time: 1.5613  data: 1.0621  max mem: 17867
[06:24:52.446891] Epoch: [168]  [2000/5004]  eta: 0:17:32  lr: 0.000077  loss: 2.3842 (2.4886)  time: 0.3538  data: 0.0002  max mem: 17867
[06:36:31.233613] Epoch: [168]  [4000/5004]  eta: 0:05:51  lr: 0.000075  loss: 2.5298 (2.4889)  time: 0.3500  data: 0.0002  max mem: 17867
[06:42:22.044505] Epoch: [168]  [5003/5004]  eta: 0:00:00  lr: 0.000074  loss: 2.4385 (2.4876)  time: 0.3431  data: 0.0011  max mem: 17867
[06:42:22.392024] Epoch: [168] Total time: 0:29:11 (0.3500 s / it)
[06:42:22.398487] Averaged stats: lr: 0.000074  loss: 2.4385 (2.4945)
[06:42:23.425899] Test:  [   0/1563]  eta: 0:26:40  loss: 0.3136 (0.3136)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.0239  data: 0.8792  max mem: 17867
[06:43:28.374753] Test:  [ 500/1563]  eta: 0:02:19  loss: 0.6912 (0.7204)  acc1: 81.2500 (82.0422)  acc5: 96.8750 (96.5070)  time: 0.1300  data: 0.0002  max mem: 17867
[06:44:33.399374] Test:  [1000/1563]  eta: 0:01:13  loss: 1.0887 (0.8441)  acc1: 65.6250 (79.2520)  acc5: 93.7500 (94.8895)  time: 0.1301  data: 0.0002  max mem: 17867
[06:45:38.356219] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5318 (0.9164)  acc1: 84.3750 (77.5275)  acc5: 96.8750 (94.0706)  time: 0.1300  data: 0.0002  max mem: 17867
[06:45:46.501879] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4755 (0.9171)  acc1: 90.6250 (77.4860)  acc5: 100.0000 (94.0860)  time: 0.1345  data: 0.0001  max mem: 17867
[06:45:46.587857] Test: Total time: 0:03:24 (0.1306 s / it)
[06:45:46.714909] * Acc@1 77.486 Acc@5 94.086 loss 0.917
[06:45:46.715090] Accuracy of the network on the 50000 test images: 77.5%
[06:45:46.715113] Max accuracy: 77.49%
[06:45:46.745560] log_dir: ./output_dir_cml_spikformer
[06:45:48.301306] Epoch: [169]  [   0/5004]  eta: 2:09:40  lr: 0.000074  loss: 2.3292 (2.3292)  time: 1.5548  data: 1.0579  max mem: 17867
[06:57:28.448542] Epoch: [169]  [2000/5004]  eta: 0:17:33  lr: 0.000072  loss: 2.5720 (2.4886)  time: 0.3489  data: 0.0002  max mem: 17867
[07:09:08.173683] Epoch: [169]  [4000/5004]  eta: 0:05:51  lr: 0.000071  loss: 2.4759 (2.4928)  time: 0.3499  data: 0.0002  max mem: 17867
[07:14:59.072233] Epoch: [169]  [5003/5004]  eta: 0:00:00  lr: 0.000070  loss: 2.4991 (2.4923)  time: 0.3460  data: 0.0014  max mem: 17867
[07:14:59.428276] Epoch: [169] Total time: 0:29:12 (0.3503 s / it)
[07:14:59.434280] Averaged stats: lr: 0.000070  loss: 2.4991 (2.4890)
[07:15:00.541477] Test:  [   0/1563]  eta: 0:28:42  loss: 0.4797 (0.4797)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.1023  data: 0.9602  max mem: 17867
[07:16:05.649117] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.6744 (0.7143)  acc1: 81.2500 (81.9673)  acc5: 96.8750 (96.7003)  time: 0.1299  data: 0.0002  max mem: 17867
[07:17:10.652119] Test:  [1000/1563]  eta: 0:01:13  loss: 1.1056 (0.8406)  acc1: 75.0000 (79.1708)  acc5: 93.7500 (94.9956)  time: 0.1300  data: 0.0002  max mem: 17867
[07:18:15.645423] Test:  [1500/1563]  eta: 0:00:08  loss: 0.4961 (0.9104)  acc1: 87.5000 (77.5400)  acc5: 96.8750 (94.0956)  time: 0.1300  data: 0.0002  max mem: 17867
[07:18:23.627452] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4632 (0.9122)  acc1: 87.5000 (77.4620)  acc5: 96.8750 (94.1100)  time: 0.1262  data: 0.0001  max mem: 17867
[07:18:23.703159] Test: Total time: 0:03:24 (0.1307 s / it)
[07:18:23.807045] * Acc@1 77.462 Acc@5 94.110 loss 0.912
[07:18:23.807205] Accuracy of the network on the 50000 test images: 77.5%
[07:18:23.807227] Max accuracy: 77.49%
[07:18:23.835402] log_dir: ./output_dir_cml_spikformer
[07:18:25.361585] Epoch: [170]  [   0/5004]  eta: 2:07:11  lr: 0.000070  loss: 3.0385 (3.0385)  time: 1.5251  data: 1.0971  max mem: 17867
[07:30:05.626271] Epoch: [170]  [2000/5004]  eta: 0:17:33  lr: 0.000068  loss: 2.3830 (2.4784)  time: 0.3509  data: 0.0006  max mem: 17867
[07:41:45.749892] Epoch: [170]  [4000/5004]  eta: 0:05:51  lr: 0.000066  loss: 2.3673 (2.4792)  time: 0.3467  data: 0.0002  max mem: 17867
[07:47:36.724662] Epoch: [170]  [5003/5004]  eta: 0:00:00  lr: 0.000065  loss: 2.4427 (2.4785)  time: 0.3452  data: 0.0006  max mem: 17867
[07:47:37.063260] Epoch: [170] Total time: 0:29:13 (0.3504 s / it)
[07:47:37.070219] Averaged stats: lr: 0.000065  loss: 2.4427 (2.4813)
[07:47:38.094030] Test:  [   0/1563]  eta: 0:26:34  loss: 0.3928 (0.3928)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.0201  data: 0.8761  max mem: 17867
[07:48:43.090136] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.6951 (0.7084)  acc1: 81.2500 (82.3416)  acc5: 96.8750 (96.6442)  time: 0.1301  data: 0.0002  max mem: 17867
[07:49:48.085446] Test:  [1000/1563]  eta: 0:01:13  loss: 1.0592 (0.8364)  acc1: 71.8750 (79.4737)  acc5: 96.8750 (95.1111)  time: 0.1299  data: 0.0002  max mem: 17867
[07:50:53.110598] Test:  [1500/1563]  eta: 0:00:08  loss: 0.4853 (0.9122)  acc1: 87.5000 (77.6795)  acc5: 96.8750 (94.1893)  time: 0.1300  data: 0.0002  max mem: 17867
[07:51:01.097261] Test:  [1562/1563]  eta: 0:00:00  loss: 0.3647 (0.9124)  acc1: 90.6250 (77.6480)  acc5: 100.0000 (94.2020)  time: 0.1262  data: 0.0001  max mem: 17867
[07:51:01.160597] Test: Total time: 0:03:24 (0.1306 s / it)
[07:51:01.446566] * Acc@1 77.648 Acc@5 94.202 loss 0.912
[07:51:01.446725] Accuracy of the network on the 50000 test images: 77.6%
[07:51:01.446751] Max accuracy: 77.65%
[07:51:01.454008] log_dir: ./output_dir_cml_spikformer
[07:51:02.960943] Epoch: [171]  [   0/5004]  eta: 2:05:36  lr: 0.000065  loss: 2.6786 (2.6786)  time: 1.5061  data: 0.9744  max mem: 17867
[08:02:44.869060] Epoch: [171]  [2000/5004]  eta: 0:17:35  lr: 0.000064  loss: 2.4253 (2.4713)  time: 0.3534  data: 0.0002  max mem: 17867
[08:14:26.123278] Epoch: [171]  [4000/5004]  eta: 0:05:52  lr: 0.000062  loss: 2.4294 (2.4734)  time: 0.3477  data: 0.0002  max mem: 17867
[08:20:17.271691] Epoch: [171]  [5003/5004]  eta: 0:00:00  lr: 0.000061  loss: 2.3692 (2.4759)  time: 0.3479  data: 0.0011  max mem: 17867
[08:20:17.662655] Epoch: [171] Total time: 0:29:16 (0.3510 s / it)
[08:20:17.669196] Averaged stats: lr: 0.000061  loss: 2.3692 (2.4759)
[08:20:18.745022] Test:  [   0/1563]  eta: 0:27:55  loss: 0.3523 (0.3523)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.0722  data: 0.9308  max mem: 17867
[08:21:23.741734] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.6691 (0.7134)  acc1: 81.2500 (82.2730)  acc5: 96.8750 (96.5382)  time: 0.1299  data: 0.0002  max mem: 17867
[08:22:28.806992] Test:  [1000/1563]  eta: 0:01:13  loss: 1.0369 (0.8394)  acc1: 71.8750 (79.2489)  acc5: 96.8750 (95.0268)  time: 0.1300  data: 0.0002  max mem: 17867
[08:23:33.811158] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5228 (0.9071)  acc1: 84.3750 (77.7086)  acc5: 96.8750 (94.2122)  time: 0.1299  data: 0.0002  max mem: 17867
[08:23:41.795429] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4465 (0.9081)  acc1: 90.6250 (77.6420)  acc5: 96.8750 (94.2240)  time: 0.1262  data: 0.0001  max mem: 17867
[08:23:41.869890] Test: Total time: 0:03:24 (0.1306 s / it)
[08:23:42.103993] * Acc@1 77.642 Acc@5 94.224 loss 0.908
[08:23:42.104162] Accuracy of the network on the 50000 test images: 77.6%
[08:23:42.104184] Max accuracy: 77.65%
[08:23:42.141159] log_dir: ./output_dir_cml_spikformer
[08:23:43.601917] Epoch: [172]  [   0/5004]  eta: 2:01:46  lr: 0.000061  loss: 2.6957 (2.6957)  time: 1.4601  data: 1.0496  max mem: 17867
[08:35:24.131676] Epoch: [172]  [2000/5004]  eta: 0:17:33  lr: 0.000059  loss: 2.4664 (2.4692)  time: 0.3506  data: 0.0002  max mem: 17867
[08:47:04.309602] Epoch: [172]  [4000/5004]  eta: 0:05:51  lr: 0.000058  loss: 2.4830 (2.4645)  time: 0.3563  data: 0.0002  max mem: 17867
[08:52:55.449552] Epoch: [172]  [5003/5004]  eta: 0:00:00  lr: 0.000057  loss: 2.4511 (2.4648)  time: 0.3459  data: 0.0006  max mem: 17867
[08:52:55.802800] Epoch: [172] Total time: 0:29:13 (0.3505 s / it)
[08:52:55.803399] Averaged stats: lr: 0.000057  loss: 2.4511 (2.4675)
[08:52:56.751802] Test:  [   0/1563]  eta: 0:24:36  loss: 0.3242 (0.3242)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 0.9449  data: 0.8078  max mem: 17867
[08:54:01.753937] Test:  [ 500/1563]  eta: 0:02:19  loss: 0.7572 (0.7121)  acc1: 81.2500 (82.2418)  acc5: 96.8750 (96.6816)  time: 0.1300  data: 0.0002  max mem: 17867
[08:55:06.766289] Test:  [1000/1563]  eta: 0:01:13  loss: 1.1136 (0.8324)  acc1: 71.8750 (79.4237)  acc5: 93.7500 (95.1267)  time: 0.1299  data: 0.0002  max mem: 17867
[08:56:11.917968] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5026 (0.9037)  acc1: 87.5000 (77.7169)  acc5: 96.8750 (94.2101)  time: 0.1304  data: 0.0002  max mem: 17867
[08:56:19.904171] Test:  [1562/1563]  eta: 0:00:00  loss: 0.3757 (0.9050)  acc1: 93.7500 (77.6840)  acc5: 100.0000 (94.2100)  time: 0.1262  data: 0.0001  max mem: 17867
[08:56:19.977328] Test: Total time: 0:03:24 (0.1306 s / it)
[08:56:20.471993] * Acc@1 77.684 Acc@5 94.210 loss 0.905
[08:56:20.472132] Accuracy of the network on the 50000 test images: 77.7%
[08:56:20.472160] Max accuracy: 77.68%
[08:56:20.478873] log_dir: ./output_dir_cml_spikformer
[08:56:22.273101] Epoch: [173]  [   0/5004]  eta: 2:29:33  lr: 0.000057  loss: 2.6093 (2.6093)  time: 1.7932  data: 1.0868  max mem: 17867
[09:08:04.285686] Epoch: [173]  [2000/5004]  eta: 0:17:36  lr: 0.000055  loss: 2.3855 (2.4584)  time: 0.3521  data: 0.0002  max mem: 17867
[09:19:45.129324] Epoch: [173]  [4000/5004]  eta: 0:05:52  lr: 0.000054  loss: 2.3502 (2.4610)  time: 0.3495  data: 0.0002  max mem: 17867
[09:25:36.735716] Epoch: [173]  [5003/5004]  eta: 0:00:00  lr: 0.000053  loss: 2.4693 (2.4629)  time: 0.3445  data: 0.0011  max mem: 17867
[09:25:37.152108] Epoch: [173] Total time: 0:29:16 (0.3511 s / it)
[09:25:37.159794] Averaged stats: lr: 0.000053  loss: 2.4693 (2.4619)
[09:25:38.227128] Test:  [   0/1563]  eta: 0:27:42  loss: 0.3280 (0.3280)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.0636  data: 0.9237  max mem: 17867
[09:26:43.257664] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.7310 (0.7096)  acc1: 81.2500 (82.0297)  acc5: 96.8750 (96.7315)  time: 0.1299  data: 0.0002  max mem: 17867
[09:27:48.270359] Test:  [1000/1563]  eta: 0:01:13  loss: 1.0051 (0.8222)  acc1: 71.8750 (79.4861)  acc5: 93.7500 (95.2516)  time: 0.1300  data: 0.0002  max mem: 17867
[09:28:53.246764] Test:  [1500/1563]  eta: 0:00:08  loss: 0.4716 (0.8918)  acc1: 87.5000 (77.8856)  acc5: 100.0000 (94.4266)  time: 0.1299  data: 0.0002  max mem: 17867
[09:29:01.334182] Test:  [1562/1563]  eta: 0:00:00  loss: 0.3954 (0.8950)  acc1: 90.6250 (77.7580)  acc5: 100.0000 (94.4260)  time: 0.1315  data: 0.0001  max mem: 17867
[09:29:01.420759] Test: Total time: 0:03:24 (0.1307 s / it)
[09:29:01.528729] * Acc@1 77.758 Acc@5 94.426 loss 0.895
[09:29:01.528868] Accuracy of the network on the 50000 test images: 77.8%
[09:29:01.528889] Max accuracy: 77.76%
[09:29:01.547447] log_dir: ./output_dir_cml_spikformer
[09:29:03.068388] Epoch: [174]  [   0/5004]  eta: 2:06:47  lr: 0.000053  loss: 2.8257 (2.8257)  time: 1.5203  data: 1.1449  max mem: 17867
[09:40:42.683353] Epoch: [174]  [2000/5004]  eta: 0:17:32  lr: 0.000051  loss: 2.4158 (2.4452)  time: 0.3460  data: 0.0002  max mem: 17867
[09:52:21.431084] Epoch: [174]  [4000/5004]  eta: 0:05:51  lr: 0.000050  loss: 2.5704 (2.4568)  time: 0.3530  data: 0.0002  max mem: 17867
[09:58:11.381401] Epoch: [174]  [5003/5004]  eta: 0:00:00  lr: 0.000049  loss: 2.3838 (2.4574)  time: 0.3447  data: 0.0006  max mem: 17867
[09:58:11.749475] Epoch: [174] Total time: 0:29:10 (0.3498 s / it)
[09:58:11.760900] Averaged stats: lr: 0.000049  loss: 2.3838 (2.4552)
[09:58:12.834318] Test:  [   0/1563]  eta: 0:27:52  loss: 0.3706 (0.3706)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.0699  data: 0.9329  max mem: 17867
[09:59:17.885592] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.6575 (0.7134)  acc1: 81.2500 (82.4476)  acc5: 96.8750 (96.7877)  time: 0.1300  data: 0.0002  max mem: 17867
[10:00:22.867008] Test:  [1000/1563]  eta: 0:01:13  loss: 1.0921 (0.8295)  acc1: 68.7500 (79.5767)  acc5: 93.7500 (95.3328)  time: 0.1303  data: 0.0002  max mem: 17867
[10:01:27.894455] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5183 (0.8977)  acc1: 87.5000 (77.9793)  acc5: 100.0000 (94.4995)  time: 0.1300  data: 0.0002  max mem: 17867
[10:01:35.875386] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4249 (0.9003)  acc1: 90.6250 (77.9060)  acc5: 100.0000 (94.4680)  time: 0.1261  data: 0.0001  max mem: 17867
[10:01:35.935236] Test: Total time: 0:03:24 (0.1306 s / it)
[10:01:36.215891] * Acc@1 77.906 Acc@5 94.468 loss 0.900
[10:01:36.216030] Accuracy of the network on the 50000 test images: 77.9%
[10:01:36.216053] Max accuracy: 77.91%
[10:01:36.243070] log_dir: ./output_dir_cml_spikformer
[10:01:38.069823] Epoch: [175]  [   0/5004]  eta: 2:32:18  lr: 0.000049  loss: 2.1210 (2.1210)  time: 1.8262  data: 0.9541  max mem: 17867
[10:13:16.069977] Epoch: [175]  [2000/5004]  eta: 0:17:30  lr: 0.000047  loss: 2.2934 (2.4451)  time: 0.3489  data: 0.0002  max mem: 17867
[10:24:53.597659] Epoch: [175]  [4000/5004]  eta: 0:05:50  lr: 0.000046  loss: 2.4290 (2.4492)  time: 0.3461  data: 0.0002  max mem: 17867
[10:30:43.397328] Epoch: [175]  [5003/5004]  eta: 0:00:00  lr: 0.000045  loss: 2.4943 (2.4514)  time: 0.3423  data: 0.0006  max mem: 17867
[10:30:43.751732] Epoch: [175] Total time: 0:29:07 (0.3492 s / it)
[10:30:43.756150] Averaged stats: lr: 0.000045  loss: 2.4943 (2.4503)
[10:30:44.808714] Test:  [   0/1563]  eta: 0:27:19  loss: 0.3061 (0.3061)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.0490  data: 0.8799  max mem: 17867
[10:31:49.792742] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.6686 (0.7156)  acc1: 78.1250 (82.3540)  acc5: 96.8750 (96.4134)  time: 0.1299  data: 0.0002  max mem: 17867
[10:32:54.793475] Test:  [1000/1563]  eta: 0:01:13  loss: 0.9205 (0.8317)  acc1: 71.8750 (79.5892)  acc5: 96.8750 (95.1049)  time: 0.1300  data: 0.0002  max mem: 17867
[10:33:59.768102] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5079 (0.9013)  acc1: 87.5000 (78.0167)  acc5: 100.0000 (94.2538)  time: 0.1299  data: 0.0002  max mem: 17867
[10:34:07.756857] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4909 (0.9029)  acc1: 87.5000 (77.9740)  acc5: 100.0000 (94.2560)  time: 0.1266  data: 0.0001  max mem: 17867
[10:34:07.821061] Test: Total time: 0:03:24 (0.1306 s / it)
[10:34:08.116621] * Acc@1 77.974 Acc@5 94.256 loss 0.903
[10:34:08.116790] Accuracy of the network on the 50000 test images: 78.0%
[10:34:08.116812] Max accuracy: 77.97%
[10:34:08.139885] log_dir: ./output_dir_cml_spikformer
[10:34:09.586230] Epoch: [176]  [   0/5004]  eta: 2:00:33  lr: 0.000045  loss: 2.6493 (2.6493)  time: 1.4456  data: 1.0993  max mem: 17867
[10:45:47.129708] Epoch: [176]  [2000/5004]  eta: 0:17:29  lr: 0.000044  loss: 2.4351 (2.4417)  time: 0.3454  data: 0.0002  max mem: 17867
[10:57:24.054306] Epoch: [176]  [4000/5004]  eta: 0:05:50  lr: 0.000042  loss: 2.2993 (2.4426)  time: 0.3465  data: 0.0002  max mem: 17867
[11:03:13.471829] Epoch: [176]  [5003/5004]  eta: 0:00:00  lr: 0.000042  loss: 2.3718 (2.4411)  time: 0.3456  data: 0.0012  max mem: 17867
[11:03:13.883423] Epoch: [176] Total time: 0:29:05 (0.3489 s / it)
[11:03:13.884081] Averaged stats: lr: 0.000042  loss: 2.3718 (2.4432)
[11:03:14.933325] Test:  [   0/1563]  eta: 0:27:14  loss: 0.3662 (0.3662)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.0457  data: 0.9036  max mem: 17867
[11:04:20.187379] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.6353 (0.7131)  acc1: 81.2500 (82.6722)  acc5: 96.8750 (96.7378)  time: 0.1299  data: 0.0002  max mem: 17867
[11:05:25.387720] Test:  [1000/1563]  eta: 0:01:13  loss: 0.9521 (0.8326)  acc1: 75.0000 (79.7609)  acc5: 93.7500 (95.2516)  time: 0.1306  data: 0.0002  max mem: 17867
[11:06:30.443089] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5629 (0.9012)  acc1: 87.5000 (78.2062)  acc5: 100.0000 (94.3933)  time: 0.1299  data: 0.0002  max mem: 17867
[11:06:38.431302] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4430 (0.9034)  acc1: 87.5000 (78.1060)  acc5: 100.0000 (94.3820)  time: 0.1263  data: 0.0001  max mem: 17867
[11:06:38.499648] Test: Total time: 0:03:24 (0.1309 s / it)
[11:06:38.783956] * Acc@1 78.106 Acc@5 94.382 loss 0.903
[11:06:38.784153] Accuracy of the network on the 50000 test images: 78.1%
[11:06:38.784181] Max accuracy: 78.11%
[11:06:38.791333] log_dir: ./output_dir_cml_spikformer
[11:06:40.900356] Epoch: [177]  [   0/5004]  eta: 2:55:51  lr: 0.000042  loss: 2.5153 (2.5153)  time: 2.1085  data: 0.9749  max mem: 17867
[11:18:20.484607] Epoch: [177]  [2000/5004]  eta: 0:17:33  lr: 0.000040  loss: 2.4705 (2.4348)  time: 0.3480  data: 0.0002  max mem: 17867
[11:29:57.420129] Epoch: [177]  [4000/5004]  eta: 0:05:50  lr: 0.000039  loss: 2.5109 (2.4350)  time: 0.3485  data: 0.0002  max mem: 17867
[11:35:47.017511] Epoch: [177]  [5003/5004]  eta: 0:00:00  lr: 0.000038  loss: 2.4038 (2.4350)  time: 0.3425  data: 0.0006  max mem: 17867
[11:35:47.353405] Epoch: [177] Total time: 0:29:08 (0.3494 s / it)
[11:35:47.356482] Averaged stats: lr: 0.000038  loss: 2.4038 (2.4393)
[11:35:48.693222] Test:  [   0/1563]  eta: 0:34:43  loss: 0.4412 (0.4412)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.3331  data: 0.9039  max mem: 17867
[11:36:53.650237] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.6261 (0.7006)  acc1: 81.2500 (82.7033)  acc5: 96.8750 (96.7378)  time: 0.1300  data: 0.0002  max mem: 17867
[11:37:58.617061] Test:  [1000/1563]  eta: 0:01:13  loss: 1.0321 (0.8201)  acc1: 71.8750 (79.9045)  acc5: 93.7500 (95.2704)  time: 0.1299  data: 0.0002  max mem: 17867
[11:39:03.600880] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5141 (0.8892)  acc1: 84.3750 (78.2062)  acc5: 100.0000 (94.4683)  time: 0.1298  data: 0.0002  max mem: 17867
[11:39:11.726526] Test:  [1562/1563]  eta: 0:00:00  loss: 0.3909 (0.8905)  acc1: 90.6250 (78.1820)  acc5: 96.8750 (94.4600)  time: 0.1335  data: 0.0001  max mem: 17867
[11:39:11.846473] Test: Total time: 0:03:24 (0.1308 s / it)
[11:39:11.998284] * Acc@1 78.182 Acc@5 94.460 loss 0.890
[11:39:11.998449] Accuracy of the network on the 50000 test images: 78.2%
[11:39:11.998471] Max accuracy: 78.18%
[11:39:12.016644] log_dir: ./output_dir_cml_spikformer
[11:39:13.764261] Epoch: [178]  [   0/5004]  eta: 2:25:40  lr: 0.000038  loss: 2.7493 (2.7493)  time: 1.7467  data: 0.9553  max mem: 17867
[11:50:50.887779] Epoch: [178]  [2000/5004]  eta: 0:17:29  lr: 0.000037  loss: 2.4431 (2.4337)  time: 0.3486  data: 0.0002  max mem: 17867
[12:02:27.156382] Epoch: [178]  [4000/5004]  eta: 0:05:50  lr: 0.000036  loss: 2.3833 (2.4350)  time: 0.3481  data: 0.0002  max mem: 17867
[12:08:16.794661] Epoch: [178]  [5003/5004]  eta: 0:00:00  lr: 0.000035  loss: 2.3982 (2.4333)  time: 0.3468  data: 0.0011  max mem: 17867
[12:08:17.174094] Epoch: [178] Total time: 0:29:05 (0.3488 s / it)
[12:08:17.174867] Averaged stats: lr: 0.000035  loss: 2.3982 (2.4307)
[12:08:18.221024] Test:  [   0/1563]  eta: 0:27:09  loss: 0.4558 (0.4558)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.0426  data: 0.9011  max mem: 17867
[12:09:23.199651] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.7196 (0.7083)  acc1: 78.1250 (82.9029)  acc5: 96.8750 (96.7378)  time: 0.1299  data: 0.0002  max mem: 17867
[12:10:28.270872] Test:  [1000/1563]  eta: 0:01:13  loss: 0.9872 (0.8261)  acc1: 71.8750 (79.9419)  acc5: 93.7500 (95.2922)  time: 0.1299  data: 0.0002  max mem: 17867
[12:11:34.467197] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5213 (0.9015)  acc1: 87.5000 (78.1750)  acc5: 100.0000 (94.4849)  time: 0.1299  data: 0.0002  max mem: 17867
[12:11:42.456210] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4170 (0.9032)  acc1: 90.6250 (78.1080)  acc5: 100.0000 (94.4720)  time: 0.1262  data: 0.0001  max mem: 17867
[12:11:42.533090] Test: Total time: 0:03:25 (0.1314 s / it)
[12:11:42.776889] * Acc@1 78.108 Acc@5 94.472 loss 0.903
[12:11:42.777037] Accuracy of the network on the 50000 test images: 78.1%
[12:11:42.777058] Max accuracy: 78.18%
[12:11:42.784069] log_dir: ./output_dir_cml_spikformer
[12:11:44.271847] Epoch: [179]  [   0/5004]  eta: 2:03:59  lr: 0.000035  loss: 2.0639 (2.0639)  time: 1.4867  data: 1.1435  max mem: 17867
[12:23:27.886591] Epoch: [179]  [2000/5004]  eta: 0:17:38  lr: 0.000034  loss: 2.3436 (2.4303)  time: 0.3459  data: 0.0002  max mem: 17867
[12:35:06.133421] Epoch: [179]  [4000/5004]  eta: 0:05:52  lr: 0.000032  loss: 2.4269 (2.4284)  time: 0.3467  data: 0.0002  max mem: 17867
[12:40:55.511197] Epoch: [179]  [5003/5004]  eta: 0:00:00  lr: 0.000032  loss: 2.4597 (2.4288)  time: 0.3457  data: 0.0006  max mem: 17867
[12:40:55.856495] Epoch: [179] Total time: 0:29:13 (0.3503 s / it)
[12:40:55.868765] Averaged stats: lr: 0.000032  loss: 2.4597 (2.4270)
[12:40:57.277986] Test:  [   0/1563]  eta: 0:36:37  loss: 0.3774 (0.3774)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.4057  data: 1.0914  max mem: 17867
[12:42:02.323800] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.5556 (0.6950)  acc1: 81.2500 (82.8780)  acc5: 96.8750 (96.6879)  time: 0.1299  data: 0.0002  max mem: 17867
[12:43:07.384143] Test:  [1000/1563]  eta: 0:01:13  loss: 0.9744 (0.8217)  acc1: 62.5000 (79.8795)  acc5: 93.7500 (95.2423)  time: 0.1299  data: 0.0002  max mem: 17867
[12:44:12.346294] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5148 (0.8891)  acc1: 84.3750 (78.1708)  acc5: 96.8750 (94.4100)  time: 0.1299  data: 0.0002  max mem: 17867
[12:44:20.581356] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4629 (0.8909)  acc1: 90.6250 (78.1140)  acc5: 100.0000 (94.4120)  time: 0.1389  data: 0.0001  max mem: 17867
[12:44:20.738074] Test: Total time: 0:03:24 (0.1311 s / it)
[12:44:20.742498] * Acc@1 78.114 Acc@5 94.412 loss 0.891
[12:44:20.742669] Accuracy of the network on the 50000 test images: 78.1%
[12:44:20.742701] Max accuracy: 78.18%
[12:44:20.767306] log_dir: ./output_dir_cml_spikformer
[12:44:22.246428] Epoch: [180]  [   0/5004]  eta: 2:03:15  lr: 0.000032  loss: 2.1364 (2.1364)  time: 1.4779  data: 1.0402  max mem: 17867
[12:56:00.498801] Epoch: [180]  [2000/5004]  eta: 0:17:30  lr: 0.000031  loss: 2.3008 (2.4267)  time: 0.3510  data: 0.0002  max mem: 17867
[13:07:38.307386] Epoch: [180]  [4000/5004]  eta: 0:05:50  lr: 0.000029  loss: 2.2327 (2.4282)  time: 0.3467  data: 0.0002  max mem: 17867
[13:13:27.761605] Epoch: [180]  [5003/5004]  eta: 0:00:00  lr: 0.000029  loss: 2.2702 (2.4285)  time: 0.3423  data: 0.0011  max mem: 17867
[13:13:28.144686] Epoch: [180] Total time: 0:29:07 (0.3492 s / it)
[13:13:28.146277] Averaged stats: lr: 0.000029  loss: 2.2702 (2.4221)
[13:13:29.175824] Test:  [   0/1563]  eta: 0:26:43  loss: 0.3115 (0.3115)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.0261  data: 0.8782  max mem: 17867
[13:14:34.194212] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.6915 (0.6952)  acc1: 81.2500 (82.6722)  acc5: 96.8750 (96.6754)  time: 0.1299  data: 0.0002  max mem: 17867
[13:15:39.168362] Test:  [1000/1563]  eta: 0:01:13  loss: 0.9239 (0.8172)  acc1: 68.7500 (79.8295)  acc5: 93.7500 (95.2579)  time: 0.1301  data: 0.0002  max mem: 17867
[13:16:44.162323] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5573 (0.8861)  acc1: 84.3750 (78.2541)  acc5: 96.8750 (94.4599)  time: 0.1299  data: 0.0002  max mem: 17867
[13:16:52.146868] Test:  [1562/1563]  eta: 0:00:00  loss: 0.3736 (0.8872)  acc1: 90.6250 (78.1920)  acc5: 100.0000 (94.4600)  time: 0.1262  data: 0.0001  max mem: 17867
[13:16:52.214798] Test: Total time: 0:03:24 (0.1306 s / it)
[13:16:52.401243] * Acc@1 78.192 Acc@5 94.460 loss 0.887
[13:16:52.401419] Accuracy of the network on the 50000 test images: 78.2%
[13:16:52.401442] Max accuracy: 78.19%
[13:16:52.407885] log_dir: ./output_dir_cml_spikformer
[13:16:53.904280] Epoch: [181]  [   0/5004]  eta: 2:04:42  lr: 0.000029  loss: 2.1712 (2.1712)  time: 1.4952  data: 1.1649  max mem: 17867
[13:28:34.161121] Epoch: [181]  [2000/5004]  eta: 0:17:33  lr: 0.000028  loss: 2.4746 (2.4227)  time: 0.3481  data: 0.0002  max mem: 17867
[13:40:11.765699] Epoch: [181]  [4000/5004]  eta: 0:05:51  lr: 0.000027  loss: 2.4261 (2.4232)  time: 0.3516  data: 0.0002  max mem: 17867
[13:46:01.700516] Epoch: [181]  [5003/5004]  eta: 0:00:00  lr: 0.000026  loss: 2.4606 (2.4203)  time: 0.3445  data: 0.0006  max mem: 17867
[13:46:02.040374] Epoch: [181] Total time: 0:29:09 (0.3496 s / it)
[13:46:02.041099] Averaged stats: lr: 0.000026  loss: 2.4606 (2.4159)
[13:46:03.071096] Test:  [   0/1563]  eta: 0:26:44  loss: 0.3953 (0.3953)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.0265  data: 0.8670  max mem: 17867
[13:47:08.162647] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.7472 (0.6976)  acc1: 81.2500 (82.7907)  acc5: 96.8750 (96.6816)  time: 0.1299  data: 0.0002  max mem: 17867
[13:48:13.224134] Test:  [1000/1563]  eta: 0:01:13  loss: 0.9808 (0.8179)  acc1: 71.8750 (79.9544)  acc5: 96.8750 (95.2110)  time: 0.1300  data: 0.0002  max mem: 17867
[13:49:18.243980] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5428 (0.8892)  acc1: 87.5000 (78.2582)  acc5: 100.0000 (94.4412)  time: 0.1299  data: 0.0002  max mem: 17867
[13:49:26.245889] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4216 (0.8915)  acc1: 90.6250 (78.1900)  acc5: 100.0000 (94.4260)  time: 0.1271  data: 0.0001  max mem: 17867
[13:49:26.355597] Test: Total time: 0:03:24 (0.1307 s / it)
[13:49:26.509130] * Acc@1 78.190 Acc@5 94.426 loss 0.891
[13:49:26.509314] Accuracy of the network on the 50000 test images: 78.2%
[13:49:26.509337] Max accuracy: 78.19%
[13:49:26.630676] log_dir: ./output_dir_cml_spikformer
[13:49:28.158689] Epoch: [182]  [   0/5004]  eta: 2:07:20  lr: 0.000026  loss: 2.2646 (2.2646)  time: 1.5269  data: 1.1705  max mem: 17867
[14:01:05.582967] Epoch: [182]  [2000/5004]  eta: 0:17:29  lr: 0.000025  loss: 2.3652 (2.4103)  time: 0.3498  data: 0.0002  max mem: 17867
[14:12:42.447573] Epoch: [182]  [4000/5004]  eta: 0:05:50  lr: 0.000024  loss: 2.3367 (2.4065)  time: 0.3475  data: 0.0002  max mem: 17867
[14:18:31.702721] Epoch: [182]  [5003/5004]  eta: 0:00:00  lr: 0.000023  loss: 2.3904 (2.4072)  time: 0.3434  data: 0.0011  max mem: 17867
[14:18:32.080751] Epoch: [182] Total time: 0:29:05 (0.3488 s / it)
[14:18:32.086960] Averaged stats: lr: 0.000023  loss: 2.3904 (2.4120)
[14:18:33.025303] Test:  [   0/1563]  eta: 0:24:20  loss: 0.4133 (0.4133)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 0.9347  data: 0.7945  max mem: 17867
[14:19:38.014429] Test:  [ 500/1563]  eta: 0:02:19  loss: 0.6674 (0.6999)  acc1: 81.2500 (83.0027)  acc5: 96.8750 (96.6255)  time: 0.1299  data: 0.0002  max mem: 17867
[14:20:43.200750] Test:  [1000/1563]  eta: 0:01:13  loss: 1.0255 (0.8176)  acc1: 71.8750 (80.1324)  acc5: 93.7500 (95.2547)  time: 0.1299  data: 0.0002  max mem: 17867
[14:21:48.176892] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5085 (0.8845)  acc1: 87.5000 (78.4435)  acc5: 100.0000 (94.5182)  time: 0.1299  data: 0.0002  max mem: 17867
[14:21:56.157461] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4338 (0.8872)  acc1: 90.6250 (78.3720)  acc5: 100.0000 (94.5160)  time: 0.1262  data: 0.0001  max mem: 17867
[14:21:56.241217] Test: Total time: 0:03:24 (0.1306 s / it)
[14:21:56.405133] * Acc@1 78.372 Acc@5 94.516 loss 0.887
[14:21:56.405297] Accuracy of the network on the 50000 test images: 78.4%
[14:21:57.543158] Max accuracy: 78.37%
[14:21:57.551048] log_dir: ./output_dir_cml_spikformer
[14:21:58.743825] Epoch: [183]  [   0/5004]  eta: 1:39:25  lr: 0.000023  loss: 2.4734 (2.4734)  time: 1.1921  data: 0.8624  max mem: 17867
[14:33:36.354424] Epoch: [183]  [2000/5004]  eta: 0:17:29  lr: 0.000022  loss: 2.2881 (2.3946)  time: 0.3462  data: 0.0002  max mem: 17867
[14:45:13.904831] Epoch: [183]  [4000/5004]  eta: 0:05:50  lr: 0.000021  loss: 2.4016 (2.4014)  time: 0.3477  data: 0.0002  max mem: 17867
[14:51:03.537090] Epoch: [183]  [5003/5004]  eta: 0:00:00  lr: 0.000021  loss: 2.3459 (2.4008)  time: 0.3437  data: 0.0011  max mem: 17867
[14:51:03.900544] Epoch: [183] Total time: 0:29:06 (0.3490 s / it)
[14:51:03.910637] Averaged stats: lr: 0.000021  loss: 2.3459 (2.4063)
[14:51:05.246406] Test:  [   0/1563]  eta: 0:34:41  loss: 0.3659 (0.3659)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.3319  data: 1.1946  max mem: 17867
[14:52:10.308538] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.8880 (0.6956)  acc1: 78.1250 (82.9466)  acc5: 96.8750 (96.8064)  time: 0.1299  data: 0.0002  max mem: 17867
[14:53:15.279332] Test:  [1000/1563]  eta: 0:01:13  loss: 0.9492 (0.8107)  acc1: 71.8750 (80.1854)  acc5: 96.8750 (95.2547)  time: 0.1299  data: 0.0002  max mem: 17867
[14:54:20.241862] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5113 (0.8788)  acc1: 84.3750 (78.4644)  acc5: 100.0000 (94.4870)  time: 0.1299  data: 0.0002  max mem: 17867
[14:54:28.223323] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4226 (0.8806)  acc1: 90.6250 (78.4080)  acc5: 100.0000 (94.4980)  time: 0.1262  data: 0.0001  max mem: 17867
[14:54:28.285726] Test: Total time: 0:03:24 (0.1308 s / it)
[14:54:28.484871] * Acc@1 78.408 Acc@5 94.498 loss 0.881
[14:54:28.485043] Accuracy of the network on the 50000 test images: 78.4%
[14:54:29.320763] Max accuracy: 78.41%
[14:54:29.471279] log_dir: ./output_dir_cml_spikformer
[14:54:30.694842] Epoch: [184]  [   0/5004]  eta: 1:41:57  lr: 0.000021  loss: 2.3348 (2.3348)  time: 1.2225  data: 0.8941  max mem: 17867
[15:06:07.478003] Epoch: [184]  [2000/5004]  eta: 0:17:27  lr: 0.000020  loss: 2.4395 (2.3940)  time: 0.3506  data: 0.0002  max mem: 17867
[15:17:44.499590] Epoch: [184]  [4000/5004]  eta: 0:05:50  lr: 0.000019  loss: 2.5986 (2.4013)  time: 0.3452  data: 0.0002  max mem: 17867
[15:23:33.224333] Epoch: [184]  [5003/5004]  eta: 0:00:00  lr: 0.000018  loss: 2.2038 (2.4023)  time: 0.3420  data: 0.0011  max mem: 17867
[15:23:33.571115] Epoch: [184] Total time: 0:29:04 (0.3485 s / it)
[15:23:33.572969] Averaged stats: lr: 0.000018  loss: 2.2038 (2.4043)
[15:23:34.610390] Test:  [   0/1563]  eta: 0:26:55  loss: 0.3293 (0.3293)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.0336  data: 0.8876  max mem: 17867
[15:24:39.587466] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.5362 (0.7014)  acc1: 81.2500 (83.0027)  acc5: 96.8750 (96.8563)  time: 0.1299  data: 0.0002  max mem: 17867
[15:25:44.819956] Test:  [1000/1563]  eta: 0:01:13  loss: 1.0003 (0.8169)  acc1: 71.8750 (80.1698)  acc5: 93.7500 (95.4296)  time: 0.1299  data: 0.0002  max mem: 17867
[15:26:49.860840] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5214 (0.8862)  acc1: 87.5000 (78.4519)  acc5: 100.0000 (94.6327)  time: 0.1299  data: 0.0002  max mem: 17867
[15:26:57.859645] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4361 (0.8887)  acc1: 93.7500 (78.3740)  acc5: 96.8750 (94.6280)  time: 0.1271  data: 0.0001  max mem: 17867
[15:26:57.923535] Test: Total time: 0:03:24 (0.1307 s / it)
[15:26:58.056660] * Acc@1 78.374 Acc@5 94.628 loss 0.889
[15:26:58.056846] Accuracy of the network on the 50000 test images: 78.4%
[15:26:58.056868] Max accuracy: 78.41%
[15:26:58.063515] log_dir: ./output_dir_cml_spikformer
[15:26:59.724951] Epoch: [185]  [   0/5004]  eta: 2:18:13  lr: 0.000018  loss: 2.3278 (2.3278)  time: 1.6573  data: 1.2672  max mem: 17867
[15:38:37.772174] Epoch: [185]  [2000/5004]  eta: 0:17:30  lr: 0.000018  loss: 2.4449 (2.3947)  time: 0.3482  data: 0.0002  max mem: 17867
[15:50:14.680664] Epoch: [185]  [4000/5004]  eta: 0:05:50  lr: 0.000017  loss: 2.4520 (2.3935)  time: 0.3477  data: 0.0002  max mem: 17867
[15:56:04.279288] Epoch: [185]  [5003/5004]  eta: 0:00:00  lr: 0.000016  loss: 2.3689 (2.3964)  time: 0.3486  data: 0.0011  max mem: 17867
[15:56:04.656279] Epoch: [185] Total time: 0:29:06 (0.3490 s / it)
[15:56:04.659220] Averaged stats: lr: 0.000016  loss: 2.3689 (2.4020)
[15:56:05.669514] Test:  [   0/1563]  eta: 0:26:13  loss: 0.3410 (0.3410)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.0066  data: 0.8693  max mem: 17867
[15:57:11.070746] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.7273 (0.6907)  acc1: 81.2500 (83.1712)  acc5: 96.8750 (96.8500)  time: 0.1299  data: 0.0002  max mem: 17867
[15:58:16.088949] Test:  [1000/1563]  eta: 0:01:13  loss: 0.9625 (0.8100)  acc1: 75.0000 (80.1573)  acc5: 93.7500 (95.3703)  time: 0.1301  data: 0.0002  max mem: 17867
[15:59:21.147896] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5622 (0.8745)  acc1: 84.3750 (78.5830)  acc5: 100.0000 (94.5869)  time: 0.1299  data: 0.0002  max mem: 17867
[15:59:29.127755] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4344 (0.8762)  acc1: 93.7500 (78.5060)  acc5: 100.0000 (94.5880)  time: 0.1262  data: 0.0001  max mem: 17867
[15:59:29.222046] Test: Total time: 0:03:24 (0.1309 s / it)
[15:59:29.240021] * Acc@1 78.506 Acc@5 94.588 loss 0.876
[15:59:29.240157] Accuracy of the network on the 50000 test images: 78.5%
[15:59:30.087209] Max accuracy: 78.51%
[15:59:30.272266] log_dir: ./output_dir_cml_spikformer
[15:59:31.544728] Epoch: [186]  [   0/5004]  eta: 1:46:02  lr: 0.000016  loss: 2.6412 (2.6412)  time: 1.2715  data: 0.9400  max mem: 17867
[16:11:08.951656] Epoch: [186]  [2000/5004]  eta: 0:17:28  lr: 0.000015  loss: 2.5259 (2.3909)  time: 0.3444  data: 0.0002  max mem: 17867
[16:22:46.258644] Epoch: [186]  [4000/5004]  eta: 0:05:50  lr: 0.000015  loss: 2.4166 (2.3983)  time: 0.3501  data: 0.0002  max mem: 17867
[16:28:35.288896] Epoch: [186]  [5003/5004]  eta: 0:00:00  lr: 0.000014  loss: 2.3448 (2.4018)  time: 0.3445  data: 0.0006  max mem: 17867
[16:28:35.686636] Epoch: [186] Total time: 0:29:05 (0.3488 s / it)
[16:28:35.696597] Averaged stats: lr: 0.000014  loss: 2.3448 (2.3982)
[16:28:36.674278] Test:  [   0/1563]  eta: 0:25:22  loss: 0.3918 (0.3918)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 0.9740  data: 0.8298  max mem: 17867
[16:29:41.743775] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.7725 (0.6905)  acc1: 81.2500 (83.0901)  acc5: 96.8750 (96.9686)  time: 0.1300  data: 0.0002  max mem: 17867
[16:30:46.783997] Test:  [1000/1563]  eta: 0:01:13  loss: 0.9643 (0.8075)  acc1: 71.8750 (80.3041)  acc5: 96.8750 (95.5544)  time: 0.1301  data: 0.0004  max mem: 17867
[16:31:51.758414] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5280 (0.8760)  acc1: 84.3750 (78.5372)  acc5: 96.8750 (94.7223)  time: 0.1299  data: 0.0002  max mem: 17867
[16:31:59.751218] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4269 (0.8781)  acc1: 90.6250 (78.4540)  acc5: 100.0000 (94.7220)  time: 0.1261  data: 0.0001  max mem: 17867
[16:31:59.826584] Test: Total time: 0:03:24 (0.1306 s / it)
[16:31:59.988642] * Acc@1 78.454 Acc@5 94.722 loss 0.878
[16:31:59.988931] Accuracy of the network on the 50000 test images: 78.5%
[16:31:59.988966] Max accuracy: 78.51%
[16:31:59.995835] log_dir: ./output_dir_cml_spikformer
[16:32:03.686182] Epoch: [187]  [   0/5004]  eta: 5:07:42  lr: 0.000014  loss: 2.1876 (2.1876)  time: 3.6896  data: 0.9841  max mem: 17867
[16:43:43.950317] Epoch: [187]  [2000/5004]  eta: 0:17:36  lr: 0.000013  loss: 2.4797 (2.3938)  time: 0.3503  data: 0.0002  max mem: 17867
[16:55:21.203704] Epoch: [187]  [4000/5004]  eta: 0:05:51  lr: 0.000013  loss: 2.4001 (2.3934)  time: 0.3483  data: 0.0002  max mem: 17867
[17:01:11.023231] Epoch: [187]  [5003/5004]  eta: 0:00:00  lr: 0.000012  loss: 2.2087 (2.3914)  time: 0.3432  data: 0.0006  max mem: 17867
[17:01:11.370553] Epoch: [187] Total time: 0:29:11 (0.3500 s / it)
[17:01:11.379890] Averaged stats: lr: 0.000012  loss: 2.2087 (2.3911)
[17:01:12.767835] Test:  [   0/1563]  eta: 0:36:01  loss: 0.3608 (0.3608)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.3830  data: 1.0593  max mem: 17867
[17:02:17.784366] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.6875 (0.6879)  acc1: 78.1250 (83.0714)  acc5: 96.8750 (96.8563)  time: 0.1299  data: 0.0002  max mem: 17867
[17:03:22.765693] Test:  [1000/1563]  eta: 0:01:13  loss: 0.9518 (0.8052)  acc1: 71.8750 (80.0824)  acc5: 93.7500 (95.4233)  time: 0.1299  data: 0.0002  max mem: 17867
[17:04:27.743507] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5180 (0.8733)  acc1: 87.5000 (78.5622)  acc5: 96.8750 (94.5745)  time: 0.1299  data: 0.0002  max mem: 17867
[17:04:35.846215] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4258 (0.8754)  acc1: 90.6250 (78.5080)  acc5: 100.0000 (94.5720)  time: 0.1323  data: 0.0001  max mem: 17867
[17:04:35.989296] Test: Total time: 0:03:24 (0.1309 s / it)
[17:04:36.068407] * Acc@1 78.508 Acc@5 94.572 loss 0.875
[17:04:36.068552] Accuracy of the network on the 50000 test images: 78.5%
[17:04:37.337996] Max accuracy: 78.51%
[17:04:37.434520] log_dir: ./output_dir_cml_spikformer
[17:04:38.511928] Epoch: [188]  [   0/5004]  eta: 1:29:48  lr: 0.000012  loss: 2.4265 (2.4265)  time: 1.0768  data: 0.7465  max mem: 17867
[17:16:17.764186] Epoch: [188]  [2000/5004]  eta: 0:17:31  lr: 0.000011  loss: 2.3322 (2.3886)  time: 0.3525  data: 0.0002  max mem: 17867
[17:27:55.865975] Epoch: [188]  [4000/5004]  eta: 0:05:50  lr: 0.000011  loss: 2.4939 (2.3862)  time: 0.3531  data: 0.0002  max mem: 17867
[17:33:46.033530] Epoch: [188]  [5003/5004]  eta: 0:00:00  lr: 0.000010  loss: 2.2580 (2.3865)  time: 0.3452  data: 0.0006  max mem: 17867
[17:33:46.397104] Epoch: [188] Total time: 0:29:08 (0.3495 s / it)
[17:33:46.402800] Averaged stats: lr: 0.000010  loss: 2.2580 (2.3901)
[17:33:47.468205] Test:  [   0/1563]  eta: 0:27:39  loss: 0.4302 (0.4302)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.0618  data: 0.9234  max mem: 17867
[17:34:52.483544] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.6502 (0.6838)  acc1: 78.1250 (83.0714)  acc5: 96.8750 (96.9873)  time: 0.1300  data: 0.0002  max mem: 17867
[17:35:57.543701] Test:  [1000/1563]  eta: 0:01:13  loss: 1.0245 (0.8052)  acc1: 71.8750 (80.3322)  acc5: 93.7500 (95.4389)  time: 0.1301  data: 0.0002  max mem: 17867
[17:37:02.516547] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5080 (0.8735)  acc1: 87.5000 (78.6892)  acc5: 96.8750 (94.6161)  time: 0.1299  data: 0.0002  max mem: 17867
[17:37:10.544423] Test:  [1562/1563]  eta: 0:00:00  loss: 0.3645 (0.8756)  acc1: 90.6250 (78.6460)  acc5: 100.0000 (94.6320)  time: 0.1275  data: 0.0001  max mem: 17867
[17:37:10.605417] Test: Total time: 0:03:24 (0.1306 s / it)
[17:37:10.983240] * Acc@1 78.646 Acc@5 94.632 loss 0.876
[17:37:10.983399] Accuracy of the network on the 50000 test images: 78.6%
[17:37:11.827932] Max accuracy: 78.65%
[17:37:11.835499] log_dir: ./output_dir_cml_spikformer
[17:37:13.185308] Epoch: [189]  [   0/5004]  eta: 1:52:29  lr: 0.000010  loss: 2.0956 (2.0956)  time: 1.3488  data: 1.0177  max mem: 17867
[17:48:52.363289] Epoch: [189]  [2000/5004]  eta: 0:17:31  lr: 0.000010  loss: 2.4025 (2.3765)  time: 0.3526  data: 0.0002  max mem: 17867
[18:00:31.013871] Epoch: [189]  [4000/5004]  eta: 0:05:51  lr: 0.000009  loss: 2.1993 (2.3757)  time: 0.3468  data: 0.0002  max mem: 17867
[18:06:20.809905] Epoch: [189]  [5003/5004]  eta: 0:00:00  lr: 0.000009  loss: 2.3562 (2.3787)  time: 0.3427  data: 0.0011  max mem: 17867
[18:06:21.179727] Epoch: [189] Total time: 0:29:09 (0.3496 s / it)
[18:06:21.217251] Averaged stats: lr: 0.000009  loss: 2.3562 (2.3866)
[18:06:22.628635] Test:  [   0/1563]  eta: 0:36:40  loss: 0.4161 (0.4161)  acc1: 90.6250 (90.6250)  acc5: 96.8750 (96.8750)  time: 1.4079  data: 0.9738  max mem: 17867
[18:07:27.733720] Test:  [ 500/1563]  eta: 0:02:21  loss: 0.6240 (0.6908)  acc1: 84.3750 (83.0901)  acc5: 96.8750 (96.8376)  time: 0.1300  data: 0.0002  max mem: 17867
[18:08:32.718684] Test:  [1000/1563]  eta: 0:01:13  loss: 1.0686 (0.8037)  acc1: 75.0000 (80.3665)  acc5: 96.8750 (95.4140)  time: 0.1299  data: 0.0002  max mem: 17867
[18:09:37.687686] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5350 (0.8723)  acc1: 84.3750 (78.7537)  acc5: 100.0000 (94.5578)  time: 0.1299  data: 0.0002  max mem: 17867
[18:09:45.670399] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4214 (0.8748)  acc1: 90.6250 (78.6760)  acc5: 100.0000 (94.5340)  time: 0.1262  data: 0.0001  max mem: 17867
[18:09:45.735865] Test: Total time: 0:03:24 (0.1308 s / it)
[18:09:45.864156] * Acc@1 78.676 Acc@5 94.534 loss 0.875
[18:09:45.864343] Accuracy of the network on the 50000 test images: 78.7%
[18:09:46.627626] Max accuracy: 78.68%
[18:09:46.636545] log_dir: ./output_dir_cml_spikformer
[18:09:48.051534] Epoch: [190]  [   0/5004]  eta: 1:57:57  lr: 0.000009  loss: 2.6786 (2.6786)  time: 1.4144  data: 1.0818  max mem: 17867
[18:21:26.778367] Epoch: [190]  [2000/5004]  eta: 0:17:31  lr: 0.000008  loss: 2.4316 (2.3880)  time: 0.3509  data: 0.0002  max mem: 17867
[18:33:03.929774] Epoch: [190]  [4000/5004]  eta: 0:05:50  lr: 0.000008  loss: 2.2408 (2.3844)  time: 0.3438  data: 0.0002  max mem: 17867
[18:38:53.533633] Epoch: [190]  [5003/5004]  eta: 0:00:00  lr: 0.000007  loss: 2.3274 (2.3855)  time: 0.3460  data: 0.0006  max mem: 17867
[18:38:53.893637] Epoch: [190] Total time: 0:29:07 (0.3492 s / it)
[18:38:53.895654] Averaged stats: lr: 0.000007  loss: 2.3274 (2.3835)
[18:38:54.913412] Test:  [   0/1563]  eta: 0:26:25  loss: 0.3569 (0.3569)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.0141  data: 0.8765  max mem: 17867
[18:39:59.966833] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.6483 (0.6920)  acc1: 81.2500 (82.9404)  acc5: 96.8750 (96.8500)  time: 0.1300  data: 0.0002  max mem: 17867
[18:41:04.941224] Test:  [1000/1563]  eta: 0:01:13  loss: 1.0386 (0.8067)  acc1: 75.0000 (80.2291)  acc5: 93.7500 (95.4702)  time: 0.1299  data: 0.0002  max mem: 17867
[18:42:09.905517] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5494 (0.8701)  acc1: 84.3750 (78.8412)  acc5: 100.0000 (94.6973)  time: 0.1299  data: 0.0002  max mem: 17867
[18:42:17.886947] Test:  [1562/1563]  eta: 0:00:00  loss: 0.3825 (0.8723)  acc1: 93.7500 (78.7640)  acc5: 100.0000 (94.6880)  time: 0.1262  data: 0.0001  max mem: 17867
[18:42:17.951801] Test: Total time: 0:03:24 (0.1306 s / it)
[18:42:18.081581] * Acc@1 78.764 Acc@5 94.688 loss 0.872
[18:42:18.081724] Accuracy of the network on the 50000 test images: 78.8%
[18:42:18.786027] Max accuracy: 78.76%
[18:42:18.811946] log_dir: ./output_dir_cml_spikformer
[18:42:20.257295] Epoch: [191]  [   0/5004]  eta: 2:00:27  lr: 0.000007  loss: 2.2301 (2.2301)  time: 1.4444  data: 1.1165  max mem: 17867
[18:54:00.734661] Epoch: [191]  [2000/5004]  eta: 0:17:33  lr: 0.000007  loss: 2.3983 (2.3903)  time: 0.3472  data: 0.0002  max mem: 17867
[19:05:37.924077] Epoch: [191]  [4000/5004]  eta: 0:05:51  lr: 0.000006  loss: 2.3759 (2.3923)  time: 0.3489  data: 0.0002  max mem: 17867
[19:11:27.429294] Epoch: [191]  [5003/5004]  eta: 0:00:00  lr: 0.000006  loss: 2.3321 (2.3899)  time: 0.3505  data: 0.0006  max mem: 17867
[19:11:27.786482] Epoch: [191] Total time: 0:29:08 (0.3495 s / it)
[19:11:27.794289] Averaged stats: lr: 0.000006  loss: 2.3321 (2.3841)
[19:11:28.794247] Test:  [   0/1563]  eta: 0:25:55  loss: 0.3771 (0.3771)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 0.9950  data: 0.8336  max mem: 17867
[19:12:33.800518] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.5929 (0.6848)  acc1: 81.2500 (83.1836)  acc5: 96.8750 (96.8688)  time: 0.1299  data: 0.0002  max mem: 17867
[19:13:38.787060] Test:  [1000/1563]  eta: 0:01:13  loss: 0.9685 (0.8044)  acc1: 71.8750 (80.3009)  acc5: 96.8750 (95.4389)  time: 0.1301  data: 0.0002  max mem: 17867
[19:14:43.808812] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5358 (0.8702)  acc1: 84.3750 (78.7038)  acc5: 96.8750 (94.6515)  time: 0.1308  data: 0.0002  max mem: 17867
[19:14:51.983652] Test:  [1562/1563]  eta: 0:00:00  loss: 0.3771 (0.8718)  acc1: 90.6250 (78.6460)  acc5: 100.0000 (94.6580)  time: 0.1342  data: 0.0001  max mem: 17867
[19:14:52.099875] Test: Total time: 0:03:24 (0.1307 s / it)
[19:14:52.200106] * Acc@1 78.646 Acc@5 94.658 loss 0.872
[19:14:52.200255] Accuracy of the network on the 50000 test images: 78.6%
[19:14:52.200278] Max accuracy: 78.76%
[19:14:52.206975] log_dir: ./output_dir_cml_spikformer
[19:14:53.711097] Epoch: [192]  [   0/5004]  eta: 2:05:18  lr: 0.000006  loss: 2.3205 (2.3205)  time: 1.5024  data: 0.9486  max mem: 17867
[19:26:31.894745] Epoch: [192]  [2000/5004]  eta: 0:17:30  lr: 0.000005  loss: 2.3140 (2.3896)  time: 0.3884  data: 0.0035  max mem: 17867
[19:38:09.568262] Epoch: [192]  [4000/5004]  eta: 0:05:50  lr: 0.000005  loss: 2.3150 (2.3831)  time: 0.3444  data: 0.0002  max mem: 17867
[19:43:59.383888] Epoch: [192]  [5003/5004]  eta: 0:00:00  lr: 0.000005  loss: 2.4147 (2.3833)  time: 0.3521  data: 0.0006  max mem: 17867
[19:43:59.778257] Epoch: [192] Total time: 0:29:07 (0.3492 s / it)
[19:43:59.779086] Averaged stats: lr: 0.000005  loss: 2.4147 (2.3808)
[19:44:00.887645] Test:  [   0/1563]  eta: 0:28:47  loss: 0.3556 (0.3556)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.1050  data: 0.9557  max mem: 17867
[19:45:05.896915] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.6411 (0.6939)  acc1: 81.2500 (83.0901)  acc5: 96.8750 (96.8563)  time: 0.1298  data: 0.0002  max mem: 17867
[19:46:10.828284] Test:  [1000/1563]  eta: 0:01:13  loss: 0.9829 (0.8120)  acc1: 71.8750 (80.3072)  acc5: 93.7500 (95.3703)  time: 0.1298  data: 0.0002  max mem: 17867
[19:47:15.775910] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5319 (0.8792)  acc1: 87.5000 (78.6601)  acc5: 100.0000 (94.6286)  time: 0.1298  data: 0.0002  max mem: 17867
[19:47:23.756609] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4335 (0.8809)  acc1: 90.6250 (78.5820)  acc5: 100.0000 (94.6340)  time: 0.1263  data: 0.0002  max mem: 17867
[19:47:23.811356] Test: Total time: 0:03:24 (0.1305 s / it)
[19:47:24.145887] * Acc@1 78.582 Acc@5 94.634 loss 0.881
[19:47:24.146035] Accuracy of the network on the 50000 test images: 78.6%
[19:47:24.146056] Max accuracy: 78.76%
[19:47:24.170949] log_dir: ./output_dir_cml_spikformer
[19:47:25.881758] Epoch: [193]  [   0/5004]  eta: 2:22:34  lr: 0.000005  loss: 2.2748 (2.2748)  time: 1.7096  data: 1.3663  max mem: 17867
[19:59:04.576398] Epoch: [193]  [2000/5004]  eta: 0:17:31  lr: 0.000004  loss: 2.4190 (2.3882)  time: 0.3474  data: 0.0002  max mem: 17867
[20:10:42.991792] Epoch: [193]  [4000/5004]  eta: 0:05:50  lr: 0.000004  loss: 2.1698 (2.3846)  time: 0.3464  data: 0.0002  max mem: 17867
[20:16:32.670502] Epoch: [193]  [5003/5004]  eta: 0:00:00  lr: 0.000004  loss: 2.4120 (2.3814)  time: 0.3480  data: 0.0011  max mem: 17867
[20:16:33.063046] Epoch: [193] Total time: 0:29:08 (0.3495 s / it)
[20:16:33.063839] Averaged stats: lr: 0.000004  loss: 2.4120 (2.3780)
[20:16:34.280114] Test:  [   0/1563]  eta: 0:31:35  loss: 0.3587 (0.3587)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.2126  data: 1.0554  max mem: 17867
[20:17:39.319746] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.6868 (0.6890)  acc1: 81.2500 (82.9778)  acc5: 96.8750 (96.7378)  time: 0.1300  data: 0.0002  max mem: 17867
[20:18:44.314555] Test:  [1000/1563]  eta: 0:01:13  loss: 0.9858 (0.8047)  acc1: 71.8750 (80.2479)  acc5: 96.8750 (95.4108)  time: 0.1299  data: 0.0002  max mem: 17867
[20:19:49.310024] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5381 (0.8723)  acc1: 87.5000 (78.6309)  acc5: 96.8750 (94.6327)  time: 0.1300  data: 0.0002  max mem: 17867
[20:19:57.300253] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4306 (0.8737)  acc1: 90.6250 (78.5660)  acc5: 96.8750 (94.6240)  time: 0.1262  data: 0.0001  max mem: 17867
[20:19:57.371999] Test: Total time: 0:03:24 (0.1307 s / it)
[20:19:57.611984] * Acc@1 78.566 Acc@5 94.624 loss 0.874
[20:19:57.612158] Accuracy of the network on the 50000 test images: 78.6%
[20:19:57.612180] Max accuracy: 78.76%
[20:19:57.619107] log_dir: ./output_dir_cml_spikformer
[20:19:59.131583] Epoch: [194]  [   0/5004]  eta: 2:06:01  lr: 0.000004  loss: 2.2729 (2.2729)  time: 1.5111  data: 0.9979  max mem: 17867
[20:31:37.839803] Epoch: [194]  [2000/5004]  eta: 0:17:31  lr: 0.000003  loss: 2.3035 (2.3835)  time: 0.3588  data: 0.0002  max mem: 17867
[20:43:16.187536] Epoch: [194]  [4000/5004]  eta: 0:05:50  lr: 0.000003  loss: 2.5027 (2.3803)  time: 0.3452  data: 0.0002  max mem: 17867
[20:49:06.178061] Epoch: [194]  [5003/5004]  eta: 0:00:00  lr: 0.000003  loss: 2.3284 (2.3781)  time: 0.3431  data: 0.0011  max mem: 17867
[20:49:06.560078] Epoch: [194] Total time: 0:29:08 (0.3495 s / it)
[20:49:06.572956] Averaged stats: lr: 0.000003  loss: 2.3284 (2.3760)
[20:49:07.670956] Test:  [   0/1563]  eta: 0:28:13  loss: 0.3649 (0.3649)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.0836  data: 0.9027  max mem: 17867
[20:50:12.660863] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.6574 (0.6880)  acc1: 81.2500 (83.0963)  acc5: 96.8750 (96.8064)  time: 0.1299  data: 0.0002  max mem: 17867
[20:51:17.632815] Test:  [1000/1563]  eta: 0:01:13  loss: 1.0067 (0.8057)  acc1: 71.8750 (80.2978)  acc5: 93.7500 (95.3297)  time: 0.1300  data: 0.0002  max mem: 17867
[20:52:22.620937] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5379 (0.8733)  acc1: 84.3750 (78.7600)  acc5: 96.8750 (94.5266)  time: 0.1299  data: 0.0002  max mem: 17867
[20:52:30.608825] Test:  [1562/1563]  eta: 0:00:00  loss: 0.3979 (0.8755)  acc1: 90.6250 (78.6780)  acc5: 100.0000 (94.5140)  time: 0.1261  data: 0.0001  max mem: 17867
[20:52:30.684279] Test: Total time: 0:03:24 (0.1306 s / it)
[20:52:30.940685] * Acc@1 78.678 Acc@5 94.514 loss 0.875
[20:52:30.940839] Accuracy of the network on the 50000 test images: 78.7%
[20:52:30.940861] Max accuracy: 78.76%
[20:52:31.088759] log_dir: ./output_dir_cml_spikformer
[20:52:32.645978] Epoch: [195]  [   0/5004]  eta: 2:09:47  lr: 0.000003  loss: 2.8136 (2.8136)  time: 1.5563  data: 1.2234  max mem: 17867
[21:04:11.258670] Epoch: [195]  [2000/5004]  eta: 0:17:31  lr: 0.000003  loss: 2.4309 (2.3741)  time: 0.3453  data: 0.0002  max mem: 17867
[21:15:49.531132] Epoch: [195]  [4000/5004]  eta: 0:05:50  lr: 0.000002  loss: 2.3613 (2.3818)  time: 0.3578  data: 0.0002  max mem: 17867
[21:21:39.837736] Epoch: [195]  [5003/5004]  eta: 0:00:00  lr: 0.000002  loss: 2.3906 (2.3799)  time: 0.3470  data: 0.0011  max mem: 17867
[21:21:40.304228] Epoch: [195] Total time: 0:29:09 (0.3496 s / it)
[21:21:40.304900] Averaged stats: lr: 0.000002  loss: 2.3906 (2.3782)
[21:21:42.148230] Test:  [   0/1563]  eta: 0:47:55  loss: 0.3596 (0.3596)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.8396  data: 1.7026  max mem: 17867
[21:22:47.252063] Test:  [ 500/1563]  eta: 0:02:22  loss: 0.6490 (0.6852)  acc1: 81.2500 (83.2023)  acc5: 96.8750 (96.7752)  time: 0.1303  data: 0.0002  max mem: 17867
[21:23:52.233678] Test:  [1000/1563]  eta: 0:01:14  loss: 1.0134 (0.8009)  acc1: 75.0000 (80.3166)  acc5: 93.7500 (95.4077)  time: 0.1299  data: 0.0002  max mem: 17867
[21:24:57.196972] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5380 (0.8671)  acc1: 84.3750 (78.6975)  acc5: 96.8750 (94.6473)  time: 0.1299  data: 0.0002  max mem: 17867
[21:25:05.176199] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4003 (0.8692)  acc1: 90.6250 (78.6340)  acc5: 100.0000 (94.6400)  time: 0.1261  data: 0.0001  max mem: 17867
[21:25:05.260038] Test: Total time: 0:03:24 (0.1311 s / it)
[21:25:05.343406] * Acc@1 78.634 Acc@5 94.640 loss 0.869
[21:25:05.343546] Accuracy of the network on the 50000 test images: 78.6%
[21:25:05.343568] Max accuracy: 78.76%
[21:25:05.350202] log_dir: ./output_dir_cml_spikformer
[21:25:06.819479] Epoch: [196]  [   0/5004]  eta: 2:02:24  lr: 0.000002  loss: 1.7668 (1.7668)  time: 1.4678  data: 1.0706  max mem: 17867
[21:36:44.323936] Epoch: [196]  [2000/5004]  eta: 0:17:29  lr: 0.000002  loss: 2.3604 (2.3690)  time: 0.3501  data: 0.0002  max mem: 17867
[21:48:21.235181] Epoch: [196]  [4000/5004]  eta: 0:05:50  lr: 0.000002  loss: 2.2332 (2.3698)  time: 0.3485  data: 0.0002  max mem: 17867
[21:54:10.210199] Epoch: [196]  [5003/5004]  eta: 0:00:00  lr: 0.000002  loss: 2.3435 (2.3707)  time: 0.3421  data: 0.0006  max mem: 17867
[21:54:10.547593] Epoch: [196] Total time: 0:29:05 (0.3488 s / it)
[21:54:10.554264] Averaged stats: lr: 0.000002  loss: 2.3435 (2.3746)
[21:54:11.663952] Test:  [   0/1563]  eta: 0:28:48  loss: 0.3917 (0.3917)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.1059  data: 0.9507  max mem: 17867
[21:55:16.788767] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.6669 (0.6787)  acc1: 81.2500 (83.5017)  acc5: 96.8750 (96.9561)  time: 0.1299  data: 0.0002  max mem: 17867
[21:56:21.818355] Test:  [1000/1563]  eta: 0:01:13  loss: 1.0033 (0.7999)  acc1: 75.0000 (80.4508)  acc5: 93.7500 (95.5170)  time: 0.1303  data: 0.0002  max mem: 17867
[21:57:26.864831] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5134 (0.8675)  acc1: 87.5000 (78.8308)  acc5: 96.8750 (94.7681)  time: 0.1299  data: 0.0002  max mem: 17867
[21:57:34.879988] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4376 (0.8691)  acc1: 90.6250 (78.7600)  acc5: 100.0000 (94.7600)  time: 0.1261  data: 0.0001  max mem: 17867
[21:57:34.941601] Test: Total time: 0:03:24 (0.1308 s / it)
[21:57:34.942657] * Acc@1 78.760 Acc@5 94.760 loss 0.869
[21:57:34.942763] Accuracy of the network on the 50000 test images: 78.8%
[21:57:34.942783] Max accuracy: 78.76%
[21:57:34.969335] log_dir: ./output_dir_cml_spikformer
[21:57:36.988890] Epoch: [197]  [   0/5004]  eta: 2:48:23  lr: 0.000002  loss: 2.7735 (2.7735)  time: 2.0190  data: 1.0932  max mem: 17867
[22:09:14.178256] Epoch: [197]  [2000/5004]  eta: 0:17:29  lr: 0.000002  loss: 2.2625 (2.3766)  time: 0.3463  data: 0.0002  max mem: 17867
[22:20:50.584457] Epoch: [197]  [4000/5004]  eta: 0:05:50  lr: 0.000001  loss: 2.3682 (2.3808)  time: 0.3460  data: 0.0002  max mem: 17867
[22:26:39.910215] Epoch: [197]  [5003/5004]  eta: 0:00:00  lr: 0.000001  loss: 2.2654 (2.3734)  time: 0.3459  data: 0.0006  max mem: 17867
[22:26:40.247639] Epoch: [197] Total time: 0:29:05 (0.3488 s / it)
[22:26:40.250002] Averaged stats: lr: 0.000001  loss: 2.2654 (2.3724)
[22:26:41.467537] Test:  [   0/1563]  eta: 0:31:37  loss: 0.3693 (0.3693)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 1.2139  data: 1.0098  max mem: 17867
[22:27:46.469085] Test:  [ 500/1563]  eta: 0:02:20  loss: 0.5948 (0.6839)  acc1: 84.3750 (83.2772)  acc5: 96.8750 (96.8875)  time: 0.1299  data: 0.0002  max mem: 17867
[22:28:51.452275] Test:  [1000/1563]  eta: 0:01:13  loss: 1.0184 (0.8016)  acc1: 65.6250 (80.4321)  acc5: 93.7500 (95.4358)  time: 0.1299  data: 0.0002  max mem: 17867
[22:29:56.560802] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5789 (0.8705)  acc1: 87.5000 (78.8870)  acc5: 100.0000 (94.6452)  time: 0.1299  data: 0.0002  max mem: 17867
[22:30:04.540934] Test:  [1562/1563]  eta: 0:00:00  loss: 0.3838 (0.8715)  acc1: 90.6250 (78.8320)  acc5: 100.0000 (94.6700)  time: 0.1262  data: 0.0001  max mem: 17867
[22:30:04.609050] Test: Total time: 0:03:24 (0.1307 s / it)
[22:30:04.676163] * Acc@1 78.832 Acc@5 94.670 loss 0.871
[22:30:04.676318] Accuracy of the network on the 50000 test images: 78.8%
[22:30:05.466654] Max accuracy: 78.83%
[22:30:05.601957] log_dir: ./output_dir_cml_spikformer
[22:30:06.908277] Epoch: [198]  [   0/5004]  eta: 1:48:53  lr: 0.000001  loss: 2.5906 (2.5906)  time: 1.3056  data: 0.9756  max mem: 17867
[22:41:45.235479] Epoch: [198]  [2000/5004]  eta: 0:17:30  lr: 0.000001  loss: 2.5361 (2.3717)  time: 0.3453  data: 0.0002  max mem: 17867
[22:53:22.088809] Epoch: [198]  [4000/5004]  eta: 0:05:50  lr: 0.000001  loss: 2.3015 (2.3768)  time: 0.3465  data: 0.0002  max mem: 17867
[22:59:11.965396] Epoch: [198]  [5003/5004]  eta: 0:00:00  lr: 0.000001  loss: 2.3028 (2.3744)  time: 0.3465  data: 0.0011  max mem: 17867
[22:59:12.356777] Epoch: [198] Total time: 0:29:06 (0.3491 s / it)
[22:59:12.360868] Averaged stats: lr: 0.000001  loss: 2.3028 (2.3723)
[22:59:13.284593] Test:  [   0/1563]  eta: 0:23:58  loss: 0.3251 (0.3251)  acc1: 96.8750 (96.8750)  acc5: 96.8750 (96.8750)  time: 0.9201  data: 0.7685  max mem: 17867
[23:00:18.253788] Test:  [ 500/1563]  eta: 0:02:19  loss: 0.6981 (0.6837)  acc1: 81.2500 (83.3271)  acc5: 96.8750 (96.9374)  time: 0.1299  data: 0.0002  max mem: 17867
[23:01:23.251518] Test:  [1000/1563]  eta: 0:01:13  loss: 1.0701 (0.8021)  acc1: 75.0000 (80.4851)  acc5: 93.7500 (95.4764)  time: 0.1307  data: 0.0002  max mem: 17867
[23:02:28.202567] Test:  [1500/1563]  eta: 0:00:08  loss: 0.5146 (0.8712)  acc1: 87.5000 (78.8932)  acc5: 100.0000 (94.6203)  time: 0.1299  data: 0.0002  max mem: 17867
[23:02:36.177733] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4023 (0.8728)  acc1: 90.6250 (78.8380)  acc5: 100.0000 (94.6220)  time: 0.1261  data: 0.0001  max mem: 17867
[23:02:36.240196] Test: Total time: 0:03:23 (0.1304 s / it)
[23:02:36.643321] * Acc@1 78.838 Acc@5 94.622 loss 0.873
[23:02:36.643466] Accuracy of the network on the 50000 test images: 78.8%
[23:02:37.048259] Max accuracy: 78.84%
[23:02:37.055062] log_dir: ./output_dir_cml_spikformer
[23:02:39.667771] Epoch: [199]  [   0/5004]  eta: 3:37:51  lr: 0.000001  loss: 2.2631 (2.2631)  time: 2.6122  data: 1.1346  max mem: 17867
[23:14:23.451579] Epoch: [199]  [2000/5004]  eta: 0:17:40  lr: 0.000001  loss: 2.3262 (2.3847)  time: 0.3556  data: 0.0002  max mem: 17867
[23:26:00.815233] Epoch: [199]  [4000/5004]  eta: 0:05:52  lr: 0.000001  loss: 2.2748 (2.3698)  time: 0.3443  data: 0.0002  max mem: 17867
[23:31:50.709309] Epoch: [199]  [5003/5004]  eta: 0:00:00  lr: 0.000001  loss: 2.2995 (2.3723)  time: 0.3423  data: 0.0006  max mem: 17867
[23:31:51.064954] Epoch: [199] Total time: 0:29:14 (0.3505 s / it)
[23:31:51.068839] Averaged stats: lr: 0.000001  loss: 2.2995 (2.3712)
[23:31:52.563303] Test:  [   0/1563]  eta: 0:38:20  loss: 0.3877 (0.3877)  acc1: 93.7500 (93.7500)  acc5: 96.8750 (96.8750)  time: 1.4717  data: 1.3275  max mem: 17867
[23:32:57.580746] Test:  [ 500/1563]  eta: 0:02:21  loss: 0.6187 (0.6822)  acc1: 81.2500 (83.4019)  acc5: 96.8750 (96.9374)  time: 0.1299  data: 0.0002  max mem: 17867
[23:34:02.627649] Test:  [1000/1563]  eta: 0:01:13  loss: 0.8921 (0.7996)  acc1: 75.0000 (80.4321)  acc5: 93.7500 (95.4858)  time: 0.1299  data: 0.0002  max mem: 17867
[23:35:07.612544] Test:  [1500/1563]  eta: 0:00:08  loss: 0.4857 (0.8662)  acc1: 87.5000 (78.9036)  acc5: 100.0000 (94.6556)  time: 0.1299  data: 0.0002  max mem: 17867
[23:35:15.670761] Test:  [1562/1563]  eta: 0:00:00  loss: 0.4058 (0.8682)  acc1: 90.6250 (78.8540)  acc5: 100.0000 (94.6540)  time: 0.1300  data: 0.0001  max mem: 17867
[23:35:15.800969] Test: Total time: 0:03:24 (0.1310 s / it)
[23:35:15.909851] * Acc@1 78.854 Acc@5 94.654 loss 0.868
[23:35:15.909990] Accuracy of the network on the 50000 test images: 78.9%
[23:35:16.561927] Max accuracy: 78.85%
[23:35:16.712475] Training time 4 days, 12:37:18
