Arch: bert-base-uncased_pt
Bs trn: 16
Bs val: 16
Hidden dim: 256
Dataset: civilcomments
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: None
Temperature: 0.05
Single pos: False
Supervised linear scale up: False
Supervised update delay: 0
Contrastive weight: 0.5
Classifier update interval: 8
Optim: AdamW
Max epoch: 5
Lr: 1e-05
Momentum: 0.9
Weight decay: 0.01
Weight decay c: 0.01
Stopping window: 30
Load encoder: 
Freeze encoder: False
Finetune epochs: 0
Clip grad norm: True
Lr scheduler classifier: 
Lr scheduler: 
Grad clip grad norm: False
Erm: False
Erm only: False
Pretrained spurious path: ./model/civilcomments/config/no_gce/JTT_sgd_no_gce_model_b_epoch0_seed10.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: 10
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: ['non_toxic', 'toxic']
Train class ratios: None
Test shift: random
Flipped: False
Q: 0.7
Pretrained bmodel: True
Cosine: False
Exp: JTT_sgd_no_gce_JTT_adamW
Tau: 1.8
Gamma: None
Remove label noise: False
Model for remove samples: 
Remove ratio: 0.03
Supervised contrast: True
Prioritize spurious pos: False
Contrastive type: cnc
Compute auroc: False
Model type: bert-base-uncased_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/civilcomments/civilcomments/config/contrastive_umaps
Log interval: 1
Log path: ./logs/civilcomments/config
Results path: ./results/civilcomments/config
Model path: ./model/civilcomments/config
Loss factor: 1
Supersample labels: False
Subsample labels: False
Weigh slice samples by loss: True
Val split: 0.1
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/CivilComments/
Target name: toxic
Confounder names: ['identities']
Image mean: 0
Image std: 0
Augment data: False
Max token length: 300
Task: civilcomments
Num classes: 2
Experiment configs: config
Experiment name: cnc-civilcomments-sw=re-na=32-np=32-nn=32-nne=0-tsr=1-t=0.05-bf=None-cw=0.5-sud=0-me=5-bst=16-o=AdamW-lr=1e-05-mo=0.9-wd=0.01-wdc=0.01-spur-me=1-bst=32-lr=0.001-mo=0.9-wd=0.0005-sts=0.2-s=10-r=0
Mi resampled: None

------------------------
> Loading spurious model
------------------------
Pretrained model loaded from ./model/civilcomments/config/no_gce/JTT_sgd_no_gce_model_b_epoch0_seed10.pt
======
# Calculate probability ...
======
