Arch: resnet50_pt
Bs trn: 256
Bs val: 256
Hidden dim: 256
Dataset: celebA
Resample class: 
Slice with: rep
Rep cluster method: gmm
Num anchor: 32
Num positive: 32
Num negative: 32
Num negative easy: 0
Weight anc by loss: False
Weight pos by loss: False
Weight neg by loss: False
Anc loss temp: 0.5
Pos loss temp: 0.5
Neg loss temp: 0.5
Data wide pos: False
Target sample ratio: 1
Balance targets: False
Additional negatives: False
Hard negative factor: 0
Full contrastive: False
Train encoder: False
No projection head: False
Projection dim: 128
Batch factor: 32
Temperature: 0.05
Single pos: False
Supervised linear scale up: False
Supervised update delay: 0
Contrastive weight: 0.5
Classifier update interval: 8
Optim: sgd
Max epoch: 50
Lr: 1e-05
Momentum: 0.9
Weight decay: 0.1
Weight decay c: 0.1
Stopping window: 30
Load encoder: 
Freeze encoder: False
Finetune epochs: 0
Clip grad norm: False
Lr scheduler classifier: 
Lr scheduler: 
Grad clip grad norm: False
Erm: False
Erm only: False
Pretrained spurious path: ./model/celebA/stage_one_erm_model_b_worst_avg_gap_best_seed0.pt
Max epoch s: 1
Bs trn s: 32
Lr s: 0.001
Momentum s: 0.9
Weight decay s: 0.0005
Slice temp: 10
Log loss interval: 10
Checkpoint interval: 50
Grad checkpoint interval: 50
Log visual interval: 100
Log grad visual interval: 50
Verbose: True
Seed: 0
Replicate: 0
No cuda: False
Resume: False
New slice: False
Num workers: 32
Evaluate: False
Data cmap: hsv
Test cmap: 
P correlation: 0.9
P corr by class: None
Train classes: ['blond', 'nonblond']
Train class ratios: None
Test shift: random
Flipped: False
Q: 0.7
Pretrained bmodel: True
Cosine: False
Exp: None
Supervised contrast: True
Prioritize spurious pos: False
Contrastive type: cnc
Compute auroc: False
Model type: resnet50_pt_cnc
Criterion: cross_entropy
Pretrained: False
Max grad norm: 1.0
Adam epsilon: 1e-08
Warmup steps: 0
Max grad norm s: 1.0
Adam epsilon s: 1e-08
Warmup steps s: 0
Grad max grad norm: 1.0
Grad adam epsilon: 1e-08
Grad warmup steps: 0
Device: cuda
Img file type: .png
Display image: False
Image path: ./images/celebA/celebA/config/contrastive_umaps
Log interval: 1
Log path: ./logs/celebA/config
Results path: ./results/celebA/config
Model path: ./model/celebA/config
Loss factor: 1
Supersample labels: False
Subsample labels: False
Weigh slice samples by loss: True
Val split: 0.2
Spurious train split: 0.2
Subsample groups: False
Train method: sc
Max robust acc: -1
Max robust epoch: -1
Max robust group acc: (None, None)
Root dir: ./datasets/data/CelebA/
Target name: Blond_Hair
Confounder names: ['Male']
Image mean: 0.449
Image std: 0.226
Augment data: False
Task: celebA
Num classes: 2
Experiment configs: config
Experiment name: cnc-celebA-sw=re-na=32-np=32-nn=32-nne=0-tsr=1-t=0.05-bf=32-cw=0.5-sud=0-me=50-bst=256-o=sgd-lr=1e-05-mo=0.9-wd=0.1-wdc=0.1-spur-me=1-bst=32-lr=0.001-mo=0.9-wd=0.0005-sts=0.2-s=0-r=0
Mi resampled: None

Loading checkpoints for train split:
[-1 -1 -1 ... -1 -1 -1]
<class 'numpy.ndarray'>
[0 1 2 3] [71629 66874 22880  1387]
Loading checkpoints for val split:
[-1 -1 -1 ... -1  1 -1]
<class 'numpy.ndarray'>
[0 1 2 3] [8535 8276 2874  182]
Loading checkpoints for test split:
[-1 -1 -1 ... -1 -1  1]
<class 'numpy.ndarray'>
[0 1 2 3] [9767 7535 2480  180]
Train dataset:
    Blond_Hair = 0, Male = 0 : n = 71629
    Blond_Hair = 0, Male = 1 : n = 66874
    Blond_Hair = 1, Male = 0 : n = 22880
    Blond_Hair = 1, Male = 1 : n = 1387
Val dataset:
    Blond_Hair = 0, Male = 0 : n = 8535
    Blond_Hair = 0, Male = 1 : n = 8276
    Blond_Hair = 1, Male = 0 : n = 2874
    Blond_Hair = 1, Male = 1 : n = 182
Test dataset:
    Blond_Hair = 0, Male = 0 : n = 9767
    Blond_Hair = 0, Male = 1 : n = 7535
    Blond_Hair = 1, Male = 0 : n = 2480
    Blond_Hair = 1, Male = 1 : n = 180
------------------------
> Loading spurious model
------------------------
Pretrained model loaded from ./model/celebA/stage_one_erm_model_b_worst_avg_gap_best_seed0.pt
======
# Calculate probability ...
======
Loading checkpoints for train split:
[-1 -1 -1 ... -1 -1 -1]
<class 'numpy.ndarray'>
[0 1 2 3] [71629 66874 22880  1387]
Loading checkpoints for val split:
[-1 -1 -1 ... -1  1 -1]
<class 'numpy.ndarray'>
[0 1 2 3] [8535 8276 2874  182]
Loading checkpoints for test split:
[-1 -1 -1 ... -1 -1  1]
<class 'numpy.ndarray'>
[0 1 2 3] [9767 7535 2480  180]
======
# Load biased model ...
======
Epoch:   1 | Train Loss: 0.000 | Train Acc: 37.407 | Val Loss: 0.001 | Val Acc: 87.069
Training:
Accuracies by groups:
0, 0  acc: 41548 / 41774 =  99.459
0, 1  acc: 14596 / 14596 = 100.000
1, 0  acc:  4738 / 95197 =   4.977
1, 1  acc:     6 / 11203 =   0.054
-------------------------------------
Average acc: 60888 / 162770 =  37.407
Robust  acc:     6 / 11203 =   0.054
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8518 /  8535 =  99.801
0, 1  acc:  8276 /  8276 = 100.000
1, 0  acc:   502 /  2874 =  17.467
1, 1  acc:     2 /   182 =   1.099
------------------------------------
Average acc: 17298 / 19867 =  87.069
Robust  acc:     2 /   182 =   1.099
------------------------------------
New max robust acc: 1.098901098901099
debias model - Saving best checkpoint at epoch 0
replace: True
-> Updating checkpoint cp-debias-wga-best-cnc-celebA-sw=re-na=32-np=32-nn=32-nne=0-tsr=1-t=0.05-bf=32-cw=0.5-sud=0-me=50-bst=256-o=sgd-lr=1e-05-mo=0.9-wd=0.1-wdc=0.1-spur-me=1-bst=32-lr=0.001-mo=0.9-wd=0.0005-sts=0.2-s=0-r=0-cpe=0-cpre=-1-cpb=-1.pt...
Checkpoint saved -------------------------------------------
Avg Test Loss: 0.001 | Avg Test Acc: 91.108
Robust Acc: 5.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 0:
Accuracies by groups:
0, 0  acc:  9739 /  9767 =  99.713
0, 1  acc: -------------------------------------------
Avg Test Loss: 0.001 | Avg Test Acc: 88.488
Robust Acc: 0.556 | Best Acc: 100.000
-------------------------------------
Training, Epoch 0:
Accuracies by groups:
0, 0  acc:  9759 /  9767 =  99.918
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:   369 /  2480 =  14.879
1, 1  acc:     1 /   180 =   0.556
------------------------------------
Average acc: 17664 / 19962 =  88.488
Robust  acc:     1 /   180 =   0.556
------------------------------------
Accuracies by groups:
0, 0  acc:  9759 /  9767 =  99.918
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:   369 /  2480 =  14.879
1, 1  acc:     1 /   180 =   0.556
------------------------------------
Average acc: 17664 / 19962 =  88.488
Robust  acc:     1 /   180 =   0.556
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9759 /  9767 =  99.918
0,Epoch:   2 | Train Loss: 0.000 | Train Acc: 20.131 | Val Loss: 0.002 | Val Acc: 87.205
Training:
Accuracies by groups:
0, 0  acc: 18677 / 22017 =  84.830
0, 1  acc:  5314 /  5385 =  98.682
1, 0  acc:  8617 / 116353 =   7.406
1, 1  acc:   159 / 19015 =Epoch:   2 | Train Loss: 0.000 | Train Acc: 38.672 | Val Loss: 0.001 | Val Acc: 88.020
Training:
Accuracies by groups:
0, 0  acc: 40946 / 41435 =  98.820
0, 1  acc: 14617 / 14620 =  99.979
1, 0  acc:  7365 / 95418 =   7.719
1, 1  acc:    19 / 11297 =   0.168
-------------------------------------
Average acc: 62947 / 162770 =  38.672
Robust  acc:    19 / 11297 =   0.168
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8513 /  8535 =  99.742
0, -------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 88.658
Robust Acc: 1.667 | Best Acc: 99.947
------------------------------------
Training, Epoch 1:
Accuracies by groups:
0, 0  acc:  9751 /  9767 =  99.836
0, 1  acc:  7531 /  7535 =  99.947
1, 0  acc:   413 /  2480 =  16.653
1, 1  acc:     3 /   180 =   1.667
------------------------------------
Average acc: 17698 / 19962 =  88.658
Robust  acc:     3 /   180 =   1.667
------------------------------------
Accuracies by groups:
0, 0  acc:  9751 /  9767 =  99.836
0, 1  acc:  7531 /  7535 =  99.947
1, 0  acc:   413 /  2480 =  16.653
1, 1  acc:     3 /   180 =   1.667
------------------------------------
Average acc: 17698 / 19962 =  88.658
Robust  acc:     3 /   180 =   1.667
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9751 /  9767 =  99.836
0, 1 -------------------------------------------
Avg Test Loss: 0.001 | Avg Test Acc: 89.215
Robust Acc: 1.667 | Best Acc: 100.000
-------------------------------------
Training, Epoch 1:
Accuracies by groups:
0, 0  acc:  9751 /  9767 =  99.836
0, 1  acEpoch:   3 | Train Loss: 0.000 | Train Acc: 18.119 | Val Loss: 0.002 | Val Acc: 85.791
Training:
Accuracies by groups:
0, 0  acc: 20472 / 21678 =  94.437
0, 1  acc:  5443 /  5496 =  99.036
1, 0  acc:  3460 / 116395 =   2.973
1, 1  acc:   118 / 19201 =   0.615
-------------------------------------
Average acc: 29493 / 162770 =  18.119
Robust  acc:   118 / 19201 =   0.615
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8523 /  8535 =  99.859
0, 1  acc:  8252 /  8276 =  99.710
1, 0  acc:   268 /  2874 =   9.325
1, 1  acc:     1 /   182 =   0.549
------------------------------------
Average acc: 17044 / 19867 =  85.791
Robust  acc:     1 /   182 =   0.549
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 87.652
Robust Acc: 0.556 | Best Acc: 99.826
------------------------------------
Training, Epoch 2:
Accuracies by groups:
0, 0  acc:  9750 /  9767 =  99.826
0, 1  acc:  7518 /  7535 =  99.774
1, 0  acc:   228 /  2480 =   9.194
1, 1  acc:     1 /   180 =   0.556
------------------------------------
Average acc: 17497 / 19962 =  87.652
Robust  acc:     1 /   180 =   0.556
------------------------------------
Accuracies by groups:
0, 0  acc:  9750 /  9767 =  99.826
0, 1  acc:  7518 /  7535 =  99.774
1, 0  acc:   228 /  2480 =   9.194
1, 1  acc:     1 /   180 =   0.556
------------------------------------
Average acc: 17497 / 19962 =  87.652
Robust  acc:     1 /   180 =   0.556
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9750 /  9767 =  99.826
0, 1  acc:  7518 /  7535 =  99.774
1, 0  acc:   228 /  2480 =   9.194
1, 1  acc:     1 /   180 =   0.556
------------------------------------
Average acc: 17497 / 19962 =  87.652
Robust  acc:     1 /   180 =   0.556
------------------------------------
Epoch:   4 | Train Loss: 0.000 | Train Acc: 17.506 | Val Loss: 0.002 | Val Acc: 85.343
Training:
Accuracies by groups:
0, 0  acc: 21160 / 21658 =  97.701
0, 1  acc:  5468 /  5484 =  99.708
1, 0  acc:  1790 / 116647 =   1.535
1, 1  acc:    76 / 18981 =   0.400
-------------------------------------
Average acc: 28494 / 162770 =  17.506
Robust  acc:    76 / 18981 =   0.400
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8531 /  8535 =  99.953
0, 1  acc:  8275 /  8276 =  99.988
1, 0  acc:   149 /  2874 =   5.184
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16955 / 19867 =  85.343
Robust  acc:     0 /   182 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 87.356
Robust Acc: 0.000 | Best Acc: 99.987
------------------------------------
Training, Epoch 3:
Accuracies by groups:
0, 0  acc:  9761 /  9767 =  99.939
0, 1  acc:  7534 /  7535 =  99.987
1, 0  acc:   143 /  2480 =   5.766
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17438 / 19962 =  87.356
Robust  acc:     0 /   180 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9761 /  9767 =  99.939
0, 1  acc:  7534 /  7535 =  99.987
1, 0  acc:   143 /  2480 =   5.766
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17438 / 19962 =  87.356
Robust  acc:     0 /   180 =   0.000Epoch:   4 | Train Loss: 0.000 | Train Acc: 44.193 | Val Loss: 0.001 | Val Acc: 90.688
Training:
Accuracies by groups:
0, 0  acc: 39615 / 41711 =  94.975
0, 1  acc: 14703 / 14730 =  99.817
1, 0  acc: 17515 / 95127 =  18.412
1, 1  acc:   100 / 11202 =   0.893
-------------------------------------
Average acc: 71933 / 162770 =  44.193
Robust  acc:   100 / 1Epoch:   5 | Train Loss: 0.000 | Train Acc: 17.042 | Val Loss: 0.002 | Val Acc: 84.683
Training:
Accuracies by groups:
0, 0  acc: 21602 / 21799 =  99.096
0, 1  acc:  5540 /  5543 =  99.946
1, 0  acc:   561 / 116320 =   0.482
1, 1  acc:    37 / 19108 =   0.194
-------------------------------------
Average acc: 27740 / 162770 =  17.042
Robust  acc:    37 / 19108 =   0.194
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8275 /  8276 =  99.988
1, 0  acc:    14 /  2874 =   0.487
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16824 / 19867 =  84.683
Robust  acc:     0 /   182 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.770
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 4:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7534 /  7535 =  99.987
1, 0  acc:    20 /  2480 =   0.806
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17321 / 19962 =  86.770
Robust  acc:     0 /   180 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7534 /  7535 =  99.987
1, 0  acc:    20 /  2480 =   0.806
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17321 / 19962 =  86.770
Robust  acc:     0 /   180 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7534 /  7535 =  99.987
1, 0  acc:    20 /  2480 =   0.806
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17321 / 19962 =  86.770
Robust  acc:     0 /   180 =   0.000
------------------------------------
Epoch:   6 | Train Loss: 0.000 | Train Acc: 16.816 | Val Loss: 0.002 | Val Acc: 84.603
Training:
Accuracies by groups:
0, 0  acc: 21969 / 21972 =  99.986
0, 1  acc:  5384 /  5384 = 100.000
1, 0  acc:    16 / 116044 =   0.014
1, 1  acc:     2 / 19370 =   0.010
-------------------------------------
Average acc: 27371 / 162770 =  16.816
Robust  acc:     2 / 19370 =   0.010
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8534 /  8535 =  99.988
0, 1  acc:  8274 /  8276 =  99.976
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16808 / 19867 =  84.603
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.660
Robust Acc: 0.000 | Best Acc: 99.987
------------------------------------
Training, Epoch 5:
Accuracies by groups:
0, 0  acc:  9765 /  9767 =  99.980
0, 1  acc:  7534 /  7535 =  99.987
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17299 / 19962 =  86.660
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9765 /  9767 =  99.980
0, 1  acc:  7534 /  7535 =  99.987
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17299 / 19962 =  86.660
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9765 /  9767 =  99.980
0, 1  acc:  7534 /  7535 =  99.987
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17299 / 19962 =  86.660
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:   7 | Train Loss: 0.000 | Train Acc: 16.737 | Val Loss: 0.002 | Val Acc: 84.613
Training:
Accuracies by groups:
0, 0  acc: 21736 / 21736 = 100.000
0, 1  acc:  5507 /  5507 = 100.000
1, 0  acc:     0 / 116265 =   0.000
1, 1  acc:     0 / 19262 =   0.000
-------------------------------------
Average acc: 27243 / 162770 =  16.737
Robust  acc:     0 / 116265 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8534 /  8535 =  99.988
0, 1  acc:  8276 /  8276 = 100.000
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16810 / 19867 =  84.613
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 6:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:   8 | Train Loss: 0.000 | Train Acc: 16.661 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 21623 / 21623 = 100.000
0, 1  acc:  5496 /  5496 = 100.000
1, 0  acc:     0 / 116709 =   0.000
1, 1  acc:     0 / 18942 =   0.000
-------------------------------------
Average acc: 27119 / 162770 =  16.661
Robust  acc:     0 / 116709 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8276 /  8276 = 100.000
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 7:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:   9 | Train Loss: 0.000 | Train Acc: 16.721 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 21677 / 21677 = 100.000
0, 1  acc:  5540 /  5540 = 100.000
1, 0  acc:     0 / 116490 =   0.000
1, 1  acc:     0 / 19063 =   0.000
-------------------------------------
Average acc: 27217 / 162770 =  16.721
Robust  acc:     0 / 116490 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8276 /  8276 = 100.000
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 8:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  10 | Train Loss: 0.000 | Train Acc: 16.753 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 21769 / 21769 = 100.000
0, 1  acc:  5500 /  5500 = 100.000
1, 0  acc:     0 / 116290 =   0.000
1, 1  acc:     0 / 19211 =   0.000
-------------------------------------
Average acc: 27269 / 162770 =  16.753
Robust  acc:     0 / 116290 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8276 /  8276 = 100.000
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 9:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  11 | Train Loss: 0.000 | Train Acc: 16.817 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 21982 / 21982 = 100.000
0, 1  acc:  5391 /  5391 = 100.000
1, 0  acc:     0 / 116256 =   0.000
1, 1  acc:     0 / 19141 =   0.000
-------------------------------------
Average acc: 27373 / 162770 =  16.817
Robust  acc:     0 / 116256 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8276 /  8276 = 100.000
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 10:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  12 | Train Loss: 0.000 | Train Acc: 16.834 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 21885 / 21885 = 100.000
0, 1  acc:  5515 /  5515 = 100.000
1, 0  acc:     0 / 116130 =   0.000
1, 1  acc:     0 / 19240 =   0.000
-------------------------------------
Average acc: 27400 / 162770 =  16.834
Robust  acc:     0 / 116130 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8276 /  8276 = 100.000
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 11:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  13 | Train Loss: 0.000 | Train Acc: 16.598 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 21576 / 21576 = 100.000
0, 1  acc:  5440 /  5440 = 100.000
1, 0  acc:     0 / 116525 =   0.000
1, 1  acc:     0 / 19229 =   0.000
-------------------------------------
Average acc: 27016 / 162770 =  16.598
Robust  acc:     0 / 116525 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8276 /  8276 = 100.000
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 12:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  14 | Train Loss: 0.000 | Train Acc: 16.874 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 21975 / 21975 = 100.000
0, 1  acc:  5491 /  5491 = 100.000
1, 0  acc:     0 / 116167 =   0.000
1, 1  acc:     0 / 19137 =   0.000
-------------------------------------
Average acc: 27466 / 162770 =  16.874
Robust  acc:     0 / 116167 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8276 /  8276 = 100.000
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 13:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  15 | Train Loss: 0.000 | Train Acc: 16.818 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 21921 / 21921 = 100.000
0, 1  acc:  5454 /  5454 = 100.000
1, 0  acc:     0 / 116397 =   0.000
1, 1  acc:     0 / 18998 =   0.000
-------------------------------------
Average acc: 27375 / 162770 =  16.818
Robust  acc:     0 / 116397 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8276 /  8276 = 100.000
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 14:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  16 | Train Loss: 0.000 | Train Acc: 16.834 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 21985 / 21985 = 100.000
0, 1  acc:  5415 /  5415 = 100.000
1, 0  acc:     0 / 116269 =   0.000
1, 1  acc:     0 / 19101 =   0.000
-------------------------------------
Average acc: 27400 / 162770 =  16.834
Robust  acc:     0 / 116269 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8276 /  8276 = 100.000
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 15:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  17 | Train Loss: 0.000 | Train Acc: 16.713 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 21773 / 21773 = 100.000
0, 1  acc:  5431 /  5431 = 100.000
1, 0  acc:     0 / 116558 =   0.000
1, 1  acc:     0 / 19008 =   0.000
-------------------------------------
Average acc: 27204 / 162770 =  16.713
Robust  acc:     0 / 116558 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8276 /  8276 = 100.000
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 16:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  18 | Train Loss: 0.000 | Train Acc: 16.646 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 21785 / 21785 = 100.000
0, 1  acc:  5309 /  5309 = 100.000
1, 0  acc:     0 / 116545 =   0.000
1, 1  acc:     0 / 19131 =   0.000
-------------------------------------
Average acc: 27094 / 162770 =  16.646
Robust  acc:     0 / 116545 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8276 /  8276 = 100.000
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 17:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  19 | Train Loss: 0.000 | Train Acc: 16.784 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 21938 / 21938 = 100.000
0, 1  acc:  5381 /  5381 = 100.000
1, 0  acc:     0 / 116225 =   0.000
1, 1  acc:     0 / 19226 =   0.000
-------------------------------------
Average acc: 27319 / 162770 =  16.784
Robust  acc:     0 / 116225 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8276 /  8276 = 100.000
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 18:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  20 | Train Loss: 0.000 | Train Acc: 16.701 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 21635 / 21635 = 100.000
0, 1  acc:  5549 /  5549 = 100.000
1, 0  acc:     0 / 116260 =   0.000
1, 1  acc:     0 / 19326 =   0.000
-------------------------------------
Average acc: 27184 / 162770 =  16.701
Robust  acc:     0 / 116260 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8276 /  8276 = 100.000
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 19:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  21 | Train Loss: 0.000 | Train Acc: 16.637 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 21765 / 21765 = 100.000
0, 1  acc:  5315 /  5315 = 100.000
1, 0  acc:     0 / 116577 =   0.000
1, 1  acc:     0 / 19113 =   0.000
-------------------------------------
Average acc: 27080 / 162770 =  16.637
Robust  acc:     0 / 116577 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8276 /  8276 = 100.000
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 20:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  22 | Train Loss: 0.000 | Train Acc: 16.806 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 21946 / 21946 = 100.000
0, 1  acc:  5409 /  5409 = 100.000
1, 0  acc:     0 / 116383 =   0.000
1, 1  acc:     0 / 19032 =   0.000
-------------------------------------
Average acc: 27355 / 162770 =  16.806
Robust  acc:     0 / 116383 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8276 /  8276 = 100.000
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 21:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  23 | Train Loss: 0.000 | Train Acc: 16.784 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 21827 / 21827 = 100.000
0, 1  acc:  5493 /  5493 = 100.000
1, 0  acc:     0 / 116388 =   0.000
1, 1  acc:     0 / 19062 =   0.000
-------------------------------------
Average acc: 27320 / 162770 =  16.784
Robust  acc:     0 / 116388 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8276 /  8276 = 100.000
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 22:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  24 | Train Loss: 0.000 | Train Acc: 16.723 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 21737 / 21737 = 100.000
0, 1  acc:  5483 /  5483 = 100.000
1, 0  acc:     0 / 116355 =   0.000
1, 1  acc:     0 / 19195 =   0.000
-------------------------------------
Average acc: 27220 / 162770 =  16.723
Robust  acc:     0 / 116355 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8276 /  8276 = 100.000
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 23:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  25 | Train Loss: 0.000 | Train Acc: 16.846 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 21929 / 21929 = 100.000
0, 1  acc:  5492 /  5492 = 100.000
1, 0  acc:     0 / 116321 =   0.000
1, 1  acc:     0 / 19028 =   0.000
-------------------------------------
Average acc: 27421 / 162770 =  16.846
Robust  acc:     0 / 116321 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8276 /  8276 = 100.000
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 24:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  26 | Train Loss: 0.000 | Train Acc: 16.661 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 21564 / 21564 = 100.000
0, 1  acc:  5555 /  5555 = 100.000
1, 0  acc:     0 / 116577 =   0.000
1, 1  acc:     0 / 19074 =   0.000
-------------------------------------
Average acc: 27119 / 162770 =  16.661
Robust  acc:     0 / 116577 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8276 /  8276 = 100.000
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 25:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  27 | Train Loss: 0.000 | Train Acc: 16.867 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 22087 / 22087 = 100.000
0, 1  acc:  5367 /  5367 = 100.000
1, 0  acc:     0 / 116332 =   0.000
1, 1  acc:     0 / 18984 =   0.000
-------------------------------------
Average acc: 27454 / 162770 =  16.867
Robust  acc:     0 / 116332 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8276 /  8276 = 100.000
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 26:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  28 | Train Loss: 0.000 | Train Acc: 16.737 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 21751 / 21751 = 100.000
0, 1  acc:  5492 /  5492 = 100.000
1, 0  acc:     0 / 116412 =   0.000
1, 1  acc:     0 / 19115 =   0.000
-------------------------------------
Average acc: 27243 / 162770 =  16.737
Robust  acc:     0 / 116412 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8276 /  8276 = 100.000
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 27:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  29 | Train Loss: 0.000 | Train Acc: 16.620 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 21664 / 21664 = 100.000
0, 1  acc:  5389 /  5389 = 100.000
1, 0  acc:     0 / 116760 =   0.000
1, 1  acc:     0 / 18957 =   0.000
-------------------------------------
Average acc: 27053 / 162770 =  16.620
Robust  acc:     0 / 116760 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8276 /  8276 = 100.000
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 28:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  30 | Train Loss: 0.000 | Train Acc: 16.881 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 21952 / 21952 = 100.000
0, 1  acc:  5525 /  5525 = 100.000
1, 0  acc:     0 / 116057 =   0.000
1, 1  acc:     0 / 19236 =   0.000
-------------------------------------
Average acc: 27477 / 162770 =  16.881
Robust  acc:     0 / 116057 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8276 /  8276 = 100.000
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 29:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  31 | Train Loss: 0.000 | Train Acc: 16.604 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 21645 / 21645 = 100.000
0, 1  acc:  5381 /  5381 = 100.000
1, 0  acc:     0 / 116453 =   0.000
1, 1  acc:     0 / 19291 =   0.000
-------------------------------------
Average acc: 27026 / 162770 =  16.604
Robust  acc:     0 / 116453 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8276 /  8276 = 100.000
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 30:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  32 | Train Loss: 0.000 | Train Acc: 16.778 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 21789 / 21789 = 100.000
0, 1  acc:  5520 /  5520 = 100.000
1, 0  acc:     0 / 116331 =   0.000
1, 1  acc:     0 / 19130 =   0.000
-------------------------------------
Average acc: 27309 / 162770 =  16.778
Robust  acc:     0 / 116331 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8276 /  8276 = 100.000
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 31:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  33 | Train Loss: 0.000 | Train Acc: 16.717 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 21855 / 21855 = 100.000
0, 1  acc:  5356 /  5357 =  99.981
1, 0  acc:     0 / 116337 =   0.000
1, 1  acc:     0 / 19221 =   0.000
-------------------------------------
Average acc: 27211 / 162770 =  16.717
Robust  acc:     0 / 116337 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8276 /  8276 = 100.000
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 32:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  34 | Train Loss: 0.000 | Train Acc: 16.617 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 21691 / 21691 = 100.000
0, 1  acc:  5355 /  5355 = 100.000
1, 0  acc:     1 / 116511 =   0.001
1, 1  acc:     0 / 19213 =   0.000
-------------------------------------
Average acc: 27047 / 162770 =  16.617
Robust  acc:     0 / 19213 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8276 /  8276 = 100.000
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 33:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  35 | Train Loss: 0.000 | Train Acc: 16.673 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 21718 / 21718 = 100.000
0, 1  acc:  5415 /  5415 = 100.000
1, 0  acc:     5 / 116236 =   0.004
1, 1  acc:     0 / 19401 =   0.000
-------------------------------------
Average acc: 27138 / 162770 =  16.673
Robust  acc:     0 / 19401 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8276 /  8276 = 100.000
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 34:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  36 | Train Loss: 0.000 | Train Acc: 16.701 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 21710 / 21710 = 100.000
0, 1  acc:  5467 /  5468 =  99.982
1, 0  acc:     7 / 116387 =   0.006
1, 1  acc:     0 / 19205 =   0.000
-------------------------------------
Average acc: 27184 / 162770 =  16.701
Robust  acc:     0 / 19205 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8276 /  8276 = 100.000
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 35:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  37 | Train Loss: 0.000 | Train Acc: 16.633 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 21668 / 21668 = 100.000
0, 1  acc:  5398 /  5399 =  99.981
1, 0  acc:     8 / 116947 =   0.007
1, 1  acc:     0 / 18756 =   0.000
-------------------------------------
Average acc: 27074 / 162770 =  16.633
Robust  acc:     0 / 18756 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8276 /  8276 = 100.000
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 36:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  38 | Train Loss: 0.000 | Train Acc: 16.846 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 21964 / 21964 = 100.000
0, 1  acc:  5447 /  5447 = 100.000
1, 0  acc:     9 / 116451 =   0.008
1, 1  acc:     0 / 18908 =   0.000
-------------------------------------
Average acc: 27420 / 162770 =  16.846
Robust  acc:     0 / 18908 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8276 /  8276 = 100.000
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 37:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  39 | Train Loss: 0.000 | Train Acc: 16.777 | Val Loss: 0.002 | Val Acc: 84.613
Training:
Accuracies by groups:
0, 0  acc: 21880 / 21880 = 100.000
0, 1  acc:  5411 /  5411 = 100.000
1, 0  acc:    17 / 116310 =   0.015
1, 1  acc:     0 / 19169 =   0.000
-------------------------------------
Average acc: 27308 / 162770 =  16.777
Robust  acc:     0 / 19169 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8275 /  8276 =  99.988
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16810 / 19867 =  84.613
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 38:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  40 | Train Loss: 0.000 | Train Acc: 16.872 | Val Loss: 0.002 | Val Acc: 84.613
Training:
Accuracies by groups:
0, 0  acc: 21959 / 21959 = 100.000
0, 1  acc:  5493 /  5493 = 100.000
1, 0  acc:    10 / 116404 =   0.009
1, 1  acc:     0 / 18914 =   0.000
-------------------------------------
Average acc: 27462 / 162770 =  16.872
Robust  acc:     0 / 18914 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8275 /  8276 =  99.988
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16810 / 19867 =  84.613
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 39:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  41 | Train Loss: 0.000 | Train Acc: 16.805 | Val Loss: 0.002 | Val Acc: 84.613
Training:
Accuracies by groups:
0, 0  acc: 21938 / 21938 = 100.000
0, 1  acc:  5394 /  5394 = 100.000
1, 0  acc:    22 / 116441 =   0.019
1, 1  acc:     0 / 18997 =   0.000
-------------------------------------
Average acc: 27354 / 162770 =  16.805
Robust  acc:     0 / 18997 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8275 /  8276 =  99.988
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16810 / 19867 =  84.613
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 40:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  42 | Train Loss: 0.000 | Train Acc: 16.595 | Val Loss: 0.002 | Val Acc: 84.613
Training:
Accuracies by groups:
0, 0  acc: 21558 / 21558 = 100.000
0, 1  acc:  5438 /  5438 = 100.000
1, 0  acc:    16 / 116828 =   0.014
1, 1  acc:     0 / 18946 =   0.000
-------------------------------------
Average acc: 27012 / 162770 =  16.595
Robust  acc:     0 / 18946 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8275 /  8276 =  99.988
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16810 / 19867 =  84.613
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 41:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  43 | Train Loss: 0.000 | Train Acc: 16.807 | Val Loss: 0.002 | Val Acc: 84.613
Training:
Accuracies by groups:
0, 0  acc: 21765 / 21765 = 100.000
0, 1  acc:  5574 /  5575 =  99.982
1, 0  acc:    17 / 116626 =   0.015
1, 1  acc:     0 / 18804 =   0.000
-------------------------------------
Average acc: 27356 / 162770 =  16.807
Robust  acc:     0 / 18804 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8275 /  8276 =  99.988
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 16810 / 19867 =  84.613
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 42:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  44 | Train Loss: 0.000 | Train Acc: 16.867 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 22018 / 22018 = 100.000
0, 1  acc:  5422 /  5422 = 100.000
1, 0  acc:    15 / 116237 =   0.013
1, 1  acc:     0 / 19093 =   0.000
-------------------------------------
Average acc: 27455 / 162770 =  16.867
Robust  acc:     0 / 19093 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8275 /  8276 =  99.988
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     1 /   182 =   0.549
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 43:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  45 | Train Loss: 0.000 | Train Acc: 16.684 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 21608 / 21608 = 100.000
0, 1  acc:  5530 /  5530 = 100.000
1, 0  acc:    19 / 116277 =   0.016
1, 1  acc:     0 / 19355 =   0.000
-------------------------------------
Average acc: 27157 / 162770 =  16.684
Robust  acc:     0 / 19355 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8275 /  8276 =  99.988
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     1 /   182 =   0.549
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 44:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  46 | Train Loss: 0.000 | Train Acc: 16.556 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 21672 / 21672 = 100.000
0, 1  acc:  5266 /  5266 = 100.000
1, 0  acc:    11 / 116810 =   0.009
1, 1  acc:     0 / 19022 =   0.000
-------------------------------------
Average acc: 26949 / 162770 =  16.556
Robust  acc:     0 / 19022 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8275 /  8276 =  99.988
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     1 /   182 =   0.549
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 45:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  47 | Train Loss: 0.000 | Train Acc: 16.867 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 21938 / 21938 = 100.000
0, 1  acc:  5500 /  5500 = 100.000
1, 0  acc:    16 / 116207 =   0.014
1, 1  acc:     0 / 19125 =   0.000
-------------------------------------
Average acc: 27454 / 162770 =  16.867
Robust  acc:     0 / 19125 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8275 /  8276 =  99.988
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     1 /   182 =   0.549
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 46:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  48 | Train Loss: 0.000 | Train Acc: 16.689 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 21709 / 21709 = 100.000
0, 1  acc:  5433 /  5433 = 100.000
1, 0  acc:    22 / 116281 =   0.019
1, 1  acc:     0 / 19347 =   0.000
-------------------------------------
Average acc: 27164 / 162770 =  16.689
Robust  acc:     0 / 19347 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8275 /  8276 =  99.988
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     1 /   182 =   0.549
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 47:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  49 | Train Loss: 0.000 | Train Acc: 16.665 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 21666 / 21666 = 100.000
0, 1  acc:  5450 /  5451 =  99.982
1, 0  acc:     9 / 116431 =   0.008
1, 1  acc:     0 / 19222 =   0.000
-------------------------------------
Average acc: 27125 / 162770 =  16.665
Robust  acc:     0 / 19222 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8275 /  8276 =  99.988
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     1 /   182 =   0.549
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 48:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Epoch:  50 | Train Loss: 0.000 | Train Acc: 16.565 | Val Loss: 0.002 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 21596 / 21596 = 100.000
0, 1  acc:  5351 /  5351 = 100.000
1, 0  acc:    16 / 116562 =   0.014
1, 1  acc:     0 / 19261 =   0.000
-------------------------------------
Average acc: 26963 / 162770 =  16.565
Robust  acc:     0 / 19261 =   0.000
-------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8535 /  8535 = 100.000
0, 1  acc:  8275 /  8276 =  99.988
1, 0  acc:     0 /  2874 =   0.000
1, 1  acc:     1 /   182 =   0.549
------------------------------------
Average acc: 16811 / 19867 =  84.618
Robust  acc:     0 /  2874 =   0.000
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 49:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9767 /  9767 = 100.000
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:     0 /  2480 =   0.000
1, 1  acc:     0 /   180 =   0.000
------------------------------------
Average acc: 17302 / 19962 =  86.675
Robust  acc:     0 /  2480 =   0.000
------------------------------------
replace: True
Checkpoint saved at ./model/celebA/config/cp-debias-end-cnc-celebA-sw=re-na=32-np=32-nn=32-nne=0-tsr=1-t=0.05-bf=32-cw=0.5-sud=0-me=50-bst=256-o=sgd-lr=1e-05-mo=0.9-wd=0.1-wdc=0.1-spur-me=1-bst=32-lr=0.001-mo=0.9-wd=0.0005-sts=0.2-s=0-r=0-cpe=49-cpre=-1-cpb=-1.pt
:  6988 /  8276 =  84.437
1, 0  acc:  2751 /  2874 =  95.720
1, 1  acc:   141 /   182 =  77.473
------------------------------------
Average acc: 16489 / 19867 =  82.997
Robust  acc:  6609 /  8535 =  77.434
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 84.721
Robust Acc: 78.333 | Best Acc: 95.645
-------------------------------------
Training, Epoch 41:
Accuracies by groups:
0, 0  acc:  8029 /  9767 =  82.205
0, 1  acc:  6370 /  7535 =  84.539
1, 0  acc:  2372 /  2480 =  95.645
1, 1  acc:   141 /   180 =  78.333
------------------------------------
Average acc: 16912 / 19962 =  84.721
Robust  acc:   141 /   180 =  78.333
------------------------------------
Accuracies by groups:
0, 0  acc:  8029 /  9767 =  82.205
0, 1  acc:  6370 /  7535 =  84.539
1, 0  acc:  2372 /  2480 =  95.645
1, 1  acc:   141 /   180 =  78.333
------------------------------------
Average acc: 16912 / 19962 =  84.721
Robust  acc:   141 /   180 =  78.333
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  8029 /  9767 =  82.205
0, 1  acc:  6370 /  7535 =  84.539
1, 0  acc:  2372 /  2480 =  95.645
1, 1  acc:   141 /   180 =  78.333
------------------------------------
Average acc: 16912 / 19962 =  84.721
Robust  acc:   141 /   180 =  78.333
------------------------------------
Epoch:  43 | Train Loss: 0.000 | Train Acc: 74.353 | Val Loss: 0.002 | Val Acc: 82.484
Training:
Accuracies by groups:
0, 0  acc: 17827 / 41873 =  42.574
0, 1  acc:  7953 / 14607 =  54.446
1, 0  acc: 86732 / 94920 =  91.374
1, 1  acc:  8512 / 11370 =  74.864
--------------------------------------
Average acc: 121024 / 162770 =  74.353
Robust  acc: 17827 / 41873 =  42.574
--------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  6555 /  8535 =  76.801
0, 1  acc:  6934 /  8276 =  83.784
1, 0  acc:  2756 /  2874 =  95.894
1, 1  acc:   142 /   182 =  78.022
------------------------------------
Average acc: 16387 / 19867 =  82.484
Robust  acc:  6555 /  8535 =  76.801
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 84.230
Robust Acc: 78.889 | Best Acc: 95.847
-------------------------------------
Training, Epoch 42:
Accuracies by groups:
0, 0  acc:  7973 /  9767 =  81.632
0, 1  acc:  6322 /  7535 =  83.902
1, 0  acc:  2377 /  2480 =  95.847
1, 1  acc:   142 /   180 =  78.889
------------------------------------
Average acc: 16814 / 19962 =  84.230
Robust  acc:   142 /   180 =  78.889
------------------------------------
Accuracies by groups:
0, 0  acc:  7973 /  9767 =  81.632
0, 1  acc:  6322 /  7535 =  83.902
1, 0  acc:  2377 /  2480 =  95.847
1, 1  acc:   142 /   180 =  78.889
------------------------------------
Average acc: 16814 / 19962 =  84.230
Robust  acc:   142 /   180 =  78.889
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  7973 /  9767 =  81.632
0, 1  acc:  6322 /  7535 =  83.902
1, 0  acc:  2377 /  2480 =  95.847
1, 1  acc:   142 /   180 =  78.889
------------------------------------
Average acc: 16814 / 19962 =  84.230
Robust  acc:   142 /   180 =  78.889
------------------------------------
Epoch:  44 | Train Loss: 0.000 | Train Acc: 74.500 | Val Loss: 0.002 | Val Acc: 82.841
Training:
Accuracies by groups:
0, 0  acc: 18146 / 41968 =  43.238
0, 1  acc:  8002 / 14569 =  54.925
1, 0  acc: 86570 / 94896 =  91.226
1, 1  acc:  8546 / 11337 =  75.381
--------------------------------------
Average acc: 121264 / 162770 =  74.500
Robust  acc: 18146 / 41968 =  43.238
--------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  6611 /  8535 =  77.458
0, 1  acc:  6949 /  8276 =  83.966
1, 0  acc:  2756 /  2874 =  95.894
1, 1  acc:   142 /   182 =  78.022
------------------------------------
Average acc: 16458 / 19867 =  82.841
Robust  acc:  6611 /  8535 =  77.458
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 84.450
Robust Acc: 78.889 | Best Acc: 95.685
-------------------------------------
Training, Epoch 43:
Accuracies by groups:
0, 0  acc:  8008 /  9767 =  81.990
0, 1  acc:  6335 /  7535 =  84.074
1, 0  acc:  2373 /  2480 =  95.685
1, 1  acc:   142 /   180 =  78.889
------------------------------------
Average acc: 16858 / 19962 =  84.450
Robust  acc:   142 /   180 =  78.889
------------------------------------
Accuracies by groups:
0, 0  acc:  8008 /  9767 =  81.990
0, 1  acc:  6335 /  7535 =  84.074
1, 0  acc:  2373 /  2480 =  95.685
1, 1  acc:   142 /   180 =  78.889
------------------------------------
Average acc: 16858 / 19962 =  84.450
Robust  acc:   142 /   180 =  78.889
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  8008 /  9767 =  81.990
0, 1  acc:  6335 /  7535 =  84.074
1, 0  acc:  2373 /  2480 =  95.685
1, 1  acc:   142 /   180 =  78.889
------------------------------------
Average acc: 16858 / 19962 =  84.450
Robust  acc:   142 /   180 =  78.889
------------------------------------
Epoch:  45 | Train Loss: 0.000 | Train Acc: 74.540 | Val Loss: 0.002 | Val Acc: 82.806
Training:
Accuracies by groups:
0, 0  acc: 18276 / 41996 =  43.518
0, 1  acc:  7844 / 14425 =  54.378
1, 0  acc: 86792 / 95130 =  91.235
1, 1  acc:  8417 / 11219 =  75.025
--------------------------------------
Average acc: 121329 / 162770 =  74.540
Robust  acc: 18276 / 41996 =  43.518
--------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  6611 /  8535 =  77.458
0, 1  acc:  6944 /  8276 =  83.905
1, 0  acc:  2755 /  2874 =  95.859
1, 1  acc:   141 /   182 =  77.473
------------------------------------
Average acc: 16451 / 19867 =  82.806
Robust  acc:  6611 /  8535 =  77.458
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 84.476
Robust Acc: 78.889 | Best Acc: 95.726
-------------------------------------
Training, Epoch 44:
Accuracies by groups:
0, 0  acc:  8015 /  9767 =  82.062
0, 1  acc:  6332 /  7535 =  84.035
1, 0  acc:  2374 /  2480 =  95.726
1, 1  acc:   142 /   180 =  78.889
------------------------------------
Average acc: 16863 / 19962 =  84.476
Robust  acc:   142 /   180 =  78.889
------------------------------------
Accuracies by groups:
0, 0  acc:  8015 /  9767 =  82.062
0, 1  acc:  6332 /  7535 =  84.035
1, 0  acc:  2374 /  2480 =  95.726
1, 1  acc:   142 /   180 =  78.889
------------------------------------
Average acc: 16863 / 19962 =  84.476
Robust  acc:   142 /   180 =  78.889
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  8015 /  9767 =  82.062
0, 1  acc:  6332 /  7535 =  84.035
1, 0  acc:  2374 /  2480 =  95.726
1, 1  acc:   142 /   180 =  78.889
------------------------------------
Average acc: 16863 / 19962 =  84.476
Robust  acc:   142 /   180 =  78.889
------------------------------------
Epoch:  46 | Train Loss: 0.000 | Train Acc: 74.568 | Val Loss: 0.002 | Val Acc: 82.489
Training:
Accuracies by groups:
0, 0  acc: 18332 / 42037 =  43.609
0, 1  acc:  7904 / 14469 =  54.627
1, 0  acc: 86556 / 94944 =  91.165
1, 1  acc:  8582 / 11320 =  75.813
--------------------------------------
Average acc: 121374 / 162770 =  74.568
Robust  acc: 18332 / 42037 =  43.609
--------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  6577 /  8535 =  77.059
0, 1  acc:  6914 /  8276 =  83.543
1, 0  acc:  2755 /  2874 =  95.859
1, 1  acc:   142 /   182 =  78.022
------------------------------------
Average acc: 16388 / 19867 =  82.489
Robust  acc:  6577 /  8535 =  77.059
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 84.190
Robust Acc: 79.444 | Best Acc: 95.766
-------------------------------------
Training, Epoch 45:
Accuracies by groups:
0, 0  acc:  7985 /  9767 =  81.755
0, 1  acc:  6303 /  7535 =  83.650
1, 0  acc:  2375 /  2480 =  95.766
1, 1  acc:   143 /   180 =  79.444
------------------------------------
Average acc: 16806 / 19962 =  84.190
Robust  acc:   143 /   180 =  79.444
------------------------------------
Accuracies by groups:
0, 0  acc:  7985 /  9767 =  81.755
0, 1  acc:  6303 /  7535 =  83.650
1, 0  acc:  2375 /  2480 =  95.766
1, 1  acc:   143 /   180 =  79.444
------------------------------------
Average acc: 16806 / 19962 =  84.190
Robust  acc:   143 /   180 =  79.444
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  7985 /  9767 =  81.755
0, 1  acc:  6303 /  7535 =  83.650
1, 0  acc:  2375 /  2480 =  95.766
1, 1  acc:   143 /   180 =  79.444
------------------------------------
Average acc: 16806 / 19962 =  84.190
Robust  acc:   143 /   180 =  79.444
------------------------------------
Epoch:  47 | Train Loss: 0.000 | Train Acc: 74.677 | Val Loss: 0.002 | Val Acc: 83.551
Training:
Accuracies by groups:
0, 0  acc: 18191 / 41732 =  43.590
0, 1  acc:  7833 / 14522 =  53.939
1, 0  acc: 86905 / 95126 =  91.358
1, 1  acc:  8623 / 11390 =  75.707
--------------------------------------
Average acc: 121552 / 162770 =  74.677
Robust  acc: 18191 / 41732 =  43.590
--------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  6703 /  8535 =  78.535
0, 1  acc:  7007 /  8276 =  84.667
1, 0  acc:  2748 /  2874 =  95.616
1, 1  acc:   141 /   182 =  77.473
------------------------------------
Average acc: 16599 / 19867 =  83.551
Robust  acc:   141 /   182 =  77.473
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 84.976
Robust Acc: 78.333 | Best Acc: 95.363
-------------------------------------
Training, Epoch 46:
Accuracies by groups:
0, 0  acc:  8080 /  9767 =  82.728
0, 1  acc:  6377 /  7535 =  84.632
1, 0  acc:  2365 /  2480 =  95.363
1, 1  acc:   141 /   180 =  78.333
------------------------------------
Average acc: 16963 / 19962 =  84.976
Robust  acc:   141 /   180 =  78.333
------------------------------------
Accuracies by groups:
0, 0  acc:  8080 /  9767 =  82.728
0, 1  acc:  6377 /  7535 =  84.632
1, 0  acc:  2365 /  2480 =  95.363
1, 1  acc:   141 /   180 =  78.333
------------------------------------
Average acc: 16963 / 19962 =  84.976
Robust  acc:   141 /   180 =  78.333
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  8080 /  9767 =  82.728
0, 1  acc:  6377 /  7535 =  84.632
1, 0  acc:  2365 /  2480 =  95.363
1, 1  acc:   141 /   180 =  78.333
------------------------------------
Average acc: 16963 / 19962 =  84.976
Robust  acc:   141 /   180 =  78.333
------------------------------------
Epoch:  48 | Train Loss: 0.000 | Train Acc: 74.747 | Val Loss: 0.002 | Val Acc: 82.635
Training:
Accuracies by groups:
0, 0  acc: 18403 / 41750 =  44.079
0, 1  acc:  7866 / 14545 =  54.080
1, 0  acc: 86650 / 94940 =  91.268
1, 1  acc:  8746 / 11535 =  75.821
--------------------------------------
Average acc: 121665 / 162770 =  74.747
Robust  acc: 18403 / 41750 =  44.079
--------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  6605 /  8535 =  77.387
0, 1  acc:  6914 /  8276 =  83.543
1, 0  acc:  2756 /  2874 =  95.894
1, 1  acc:   142 /   182 =  78.022
------------------------------------
Average acc: 16417 / 19867 =  82.635
Robust  acc:  6605 /  8535 =  77.387
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 84.295
Robust Acc: 79.444 | Best Acc: 95.726
-------------------------------------
Training, Epoch 47:
Accuracies by groups:
0, 0  acc:  8008 /  9767 =  81.990
0, 1  acc:  6302 /  7535 =  83.636
1, 0  acc:  2374 /  2480 =  95.726
1, 1  acc:   143 /   180 =  79.444
------------------------------------
Average acc: 16827 / 19962 =  84.295
Robust  acc:   143 /   180 =  79.444
------------------------------------
Accuracies by groups:
0, 0  acc:  8008 /  9767 =  81.990
0, 1  acc:  6302 /  7535 =  83.636
1, 0  acc:  2374 /  2480 =  95.726
1, 1  acc:   143 /   180 =  79.444
------------------------------------
Average acc: 16827 / 19962 =  84.295
Robust  acc:   143 /   180 =  79.444
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  8008 /  9767 =  81.990
0, 1  acc:  6302 /  7535 =  83.636
1, 0  acc:  2374 /  2480 =  95.726
1, 1  acc:   143 /   180 =  79.444
------------------------------------
Average acc: 16827 / 19962 =  84.295
Robust  acc:   143 /   180 =  79.444
------------------------------------
Epoch:  49 | Train Loss: 0.000 | Train Acc: 74.931 | Val Loss: 0.002 | Val Acc: 82.408
Training:
Accuracies by groups:
0, 0  acc: 18384 / 41602 =  44.190
0, 1  acc:  7771 / 14404 =  53.950
1, 0  acc: 87186 / 95412 =  91.378
1, 1  acc:  8625 / 11352 =  75.978
--------------------------------------
Average acc: 121966 / 162770 =  74.931
Robust  acc: 18384 / 41602 =  44.190
--------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  6584 /  8535 =  77.141
0, 1  acc:  6887 /  8276 =  83.217
1, 0  acc:  2757 /  2874 =  95.929
1, 1  acc:   144 /   182 =  79.121
------------------------------------
Average acc: 16372 / 19867 =  82.408
Robust  acc:  6584 /  8535 =  77.141
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 84.095
Robust Acc: 79.444 | Best Acc: 95.847
-------------------------------------
Training, Epoch 48:
Accuracies by groups:
0, 0  acc:  7989 /  9767 =  81.796
0, 1  acc:  6278 /  7535 =  83.318
1, 0  acc:  2377 /  2480 =  95.847
1, 1  acc:   143 /   180 =  79.444
------------------------------------
Average acc: 16787 / 19962 =  84.095
Robust  acc:   143 /   180 =  79.444
------------------------------------
Accuracies by groups:
0, 0  acc:  7989 /  9767 =  81.796
0, 1  acc:  6278 /  7535 =  83.318
1, 0  acc:  2377 /  2480 =  95.847
1, 1  acc:   143 /   180 =  79.444
------------------------------------
Average acc: 16787 / 19962 =  84.095
Robust  acc:   143 /   180 =  79.444
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  7989 /  9767 =  81.796
0, 1  acc:  6278 /  7535 =  83.318
1, 0  acc:  2377 /  2480 =  95.847
1, 1  acc:   143 /   180 =  79.444
------------------------------------
Average acc: 16787 / 19962 =  84.095
Robust  acc:   143 /   180 =  79.444
------------------------------------
Epoch:  50 | Train Loss: 0.000 | Train Acc: 74.801 | Val Loss: 0.002 | Val Acc: 82.514
Training:
Accuracies by groups:
0, 0  acc: 18689 / 42099 =  44.393
0, 1  acc:  7857 / 14529 =  54.078
1, 0  acc: 86704 / 94911 =  91.353
1, 1  acc:  8504 / 11231 =  75.719
--------------------------------------
Average acc: 121754 / 162770 =  74.801
Robust  acc: 18689 / 42099 =  44.393
--------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  6593 /  8535 =  77.247
0, 1  acc:  6898 /  8276 =  83.349
1, 0  acc:  2758 /  2874 =  95.964
1, 1  acc:   144 /   182 =  79.121
------------------------------------
Average acc: 16393 / 19867 =  82.514
Robust  acc:  6593 /  8535 =  77.247
------------------------------------
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 84.170
Robust Acc: 80.000 | Best Acc: 95.766
-------------------------------------
Training, Epoch 49:
Accuracies by groups:
0, 0  acc:  7996 /  9767 =  81.868
0, 1  acc:  6287 /  7535 =  83.437
1, 0  acc:  2375 /  2480 =  95.766
1, 1  acc:   144 /   180 =  80.000
------------------------------------
Average acc: 16802 / 19962 =  84.170
Robust  acc:   144 /   180 =  80.000
------------------------------------
Accuracies by groups:
0, 0  acc:  7996 /  9767 =  81.868
0, 1  acc:  6287 /  7535 =  83.437
1, 0  acc:  2375 /  2480 =  95.766
1, 1  acc:   144 /   180 =  80.000
------------------------------------
Average acc: 16802 / 19962 =  84.170
Robust  acc:   144 /   180 =  80.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  7996 /  9767 =  81.868
0, 1  acc:  6287 /  7535 =  83.437
1, 0  acc:  2375 /  2480 =  95.766
1, 1  acc:   144 /   180 =  80.000
------------------------------------
Average acc: 16802 / 19962 =  84.170
Robust  acc:   144 /   180 =  80.000
------------------------------------
replace: True
-> Updating checkpoint cp-debias-end-cnc-celebA-sw=re-na=32-np=32-nn=32-nne=0-tsr=1-t=0.05-bf=32-cw=0.5-sud=0-me=50-bst=256-o=sgd-lr=1e-05-mo=0.9-wd=0.1-wdc=0.1-spur-me=1-bst=32-lr=0.001-mo=0.9-wd=0.0005-sts=0.2-s=0-r=0-cpe=49-cpre=-1-cpb=-1.pt...
Checkpoint saved at ./model/celebA/config/cp-debias-end-cnc-celebA-sw=re-na=32-np=32-nn=32-nne=0-tsr=1-t=0.05-bf=32-cw=0.5-sud=0-me=50-bst=256-o=sgd-lr=1e-05-mo=0.9-wd=0.1-wdc=0.1-spur-me=1-bst=32-lr=0.001-mo=0.9-wd=0.0005-sts=0.2-s=0-r=0-cpe=49-cpre=-1-cpb=-1.pt
