Arch: resnet50_pt
Bs trn: 128
Bs val: 128
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: 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: sgd
Max epoch: 5
Lr: 0.0001
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: 
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: 23
Replicate: 0
No cuda: False
Resume: False
New slice: False
Num workers: 12
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: False
Cosine: False
Exp: stage_one_erm
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=None-cw=0.5-sud=0-me=5-bst=128-o=sgd-lr=0.0001-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=23-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
Pretrained model loaded from 
Epoch:   1 | Train Loss: 0.000 | Train Acc: 84.998 | Val Loss: 0.003 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 71553 / 71629 =  99.894
0, 1  acc: 66761 / 66874 =  99.831
1, 0  acc:    35 / 22880 =   0.153
1, 1  acc:     2 /  1387 =   0.144
--------------------------------------
Average acc: 138351 / 162770 =  84.998
Robust  acc:     2 /  1387 =   0.144
--------------------------------------
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
------------------------------------
Save biased model at epoch 0
replace: True
Checkpoint saved at ./model/celebA/config/stage_one_erm_model_b_epoch0_seed23.pt
New max average-worst acc gap: 84.61770775658127
bias model - Saving best checkpoint at epoch 0
replace: True
Checkpoint saved at ./model/celebA/config/stage_one_erm_model_b_worst_avg_gap_best_epoch0_seed23.pt
-------------------------------------------
Avg Test Loss: 0.003 | Avg Test Acc: 86.675
Robust Acc: 0.000 | Best Acc: 100.000
-------------------------------------
Training, Epoch 0:
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:   2 | Train Loss: 0.001 | Train Acc: 88.080 | Val Loss: 0.001 | Val Acc: 92.374
Training:
Accuracies by groups:
0, 0  acc: 71075 / 71629 =  99.227
0, 1  acc: 66856 / 66874 =  99.973
1, 0  acc:  5411 / 22880 =  23.649
1, 1  acc:    26 /  1387 =   1.875
--------------------------------------
Average acc: 143368 / 162770 =  88.080
Robust  acc:    26 /  1387 =   1.875
--------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8247 /  8535 =  96.626
0, 1  acc:  8273 /  8276 =  99.964
1, 0  acc:  1824 /  2874 =  63.466
1, 1  acc:     8 /   182 =   4.396
------------------------------------
Average acc: 18352 / 19867 =  92.374
Robust  acc:     8 /   182 =   4.396
------------------------------------
Save biased model at epoch 1
replace: True
Checkpoint saved at ./model/celebA/config/stage_one_erm_model_b_epoch1_seed23.pt
New max average-worst acc gap: 87.97868462639188
bias model - Saving best checkpoint at epoch 1
replace: True
Checkpoint saved at ./model/celebA/config/stage_one_erm_model_b_worst_avg_gap_best_epoch1_seed23.pt
-------------------------------------------
Avg Test Loss: 0.001 | Avg Test Acc: 92.821
Robust Acc: 10.000 | Best Acc: 99.987
-------------------------------------
Training, Epoch 1:
Accuracies by groups:
0, 0  acc:  9557 /  9767 =  97.850
0, 1  acc:  7534 /  7535 =  99.987
1, 0  acc:  1420 /  2480 =  57.258
1, 1  acc:    18 /   180 =  10.000
------------------------------------
Average acc: 18529 / 19962 =  92.821
Robust  acc:    18 /   180 =  10.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9557 /  9767 =  97.850
0, 1  acc:  7534 /  7535 =  99.987
1, 0  acc:  1420 /  2480 =  57.258
1, 1  acc:    18 /   180 =  10.000
------------------------------------
Average acc: 18529 / 19962 =  92.821
Robust  acc:    18 /   180 =  10.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9557 /  9767 =  97.850
0, 1  acc:  7534 /  7535 =  99.987
1, 0  acc:  1420 /  2480 =  57.258
1, 1  acc:    18 /   180 =  10.000
------------------------------------
Average acc: 18529 / 19962 =  92.821
Robust  acc:    18 /   180 =  10.000
------------------------------------
Epoch:   3 | Train Loss: 0.000 | Train Acc: 93.377 | Val Loss: 0.001 | Val Acc: 94.081
Training:
Accuracies by groups:
0, 0  acc: 69214 / 71629 =  96.628
0, 1  acc: 66710 / 66874 =  99.755
1, 0  acc: 15840 / 22880 =  69.231
1, 1  acc:   226 /  1387 =  16.294
--------------------------------------
Average acc: 151990 / 162770 =  93.377
Robust  acc:   226 /  1387 =  16.294
--------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8114 /  8535 =  95.067
0, 1  acc:  8255 /  8276 =  99.746
1, 0  acc:  2291 /  2874 =  79.715
1, 1  acc:    31 /   182 =  17.033
------------------------------------
Average acc: 18691 / 19867 =  94.081
Robust  acc:    31 /   182 =  17.033
------------------------------------
Save biased model at epoch 2
replace: True
Checkpoint saved at ./model/celebA/config/stage_one_erm_model_b_epoch2_seed23.pt
-------------------------------------------
Avg Test Loss: 0.001 | Avg Test Acc: 94.490
Robust Acc: 23.889 | Best Acc: 99.814
-------------------------------------
Training, Epoch 2:
Accuracies by groups:
0, 0  acc:  9424 /  9767 =  96.488
0, 1  acc:  7521 /  7535 =  99.814
1, 0  acc:  1874 /  2480 =  75.565
1, 1  acc:    43 /   180 =  23.889
------------------------------------
Average acc: 18862 / 19962 =  94.490
Robust  acc:    43 /   180 =  23.889
------------------------------------
Accuracies by groups:
0, 0  acc:  9424 /  9767 =  96.488
0, 1  acc:  7521 /  7535 =  99.814
1, 0  acc:  1874 /  2480 =  75.565
1, 1  acc:    43 /   180 =  23.889
------------------------------------
Average acc: 18862 / 19962 =  94.490
Robust  acc:    43 /   180 =  23.889
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9424 /  9767 =  96.488
0, 1  acc:  7521 /  7535 =  99.814
1, 0  acc:  1874 /  2480 =  75.565
1, 1  acc:    43 /   180 =  23.889
------------------------------------
Average acc: 18862 / 19962 =  94.490
Robust  acc:    43 /   180 =  23.889
------------------------------------
Epoch:   4 | Train Loss: 0.000 | Train Acc: 94.285 | Val Loss: 0.001 | Val Acc: 94.624
Training:
Accuracies by groups:
0, 0  acc: 68877 / 71629 =  96.158
0, 1  acc: 66599 / 66874 =  99.589
1, 0  acc: 17635 / 22880 =  77.076
1, 1  acc:   357 /  1387 =  25.739
--------------------------------------
Average acc: 153468 / 162770 =  94.285
Robust  acc:   357 /  1387 =  25.739
--------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8198 /  8535 =  96.052
0, 1  acc:  8257 /  8276 =  99.770
1, 0  acc:  2309 /  2874 =  80.341
1, 1  acc:    35 /   182 =  19.231
------------------------------------
Average acc: 18799 / 19867 =  94.624
Robust  acc:    35 /   182 =  19.231
------------------------------------
Save biased model at epoch 3
replace: True
Checkpoint saved at ./model/celebA/config/stage_one_erm_model_b_epoch3_seed23.pt
-------------------------------------------
Avg Test Loss: 0.001 | Avg Test Acc: 94.860
Robust Acc: 26.111 | Best Acc: 99.774
-------------------------------------
Training, Epoch 3:
Accuracies by groups:
0, 0  acc:  9490 /  9767 =  97.164
0, 1  acc:  7518 /  7535 =  99.774
1, 0  acc:  1881 /  2480 =  75.847
1, 1  acc:    47 /   180 =  26.111
------------------------------------
Average acc: 18936 / 19962 =  94.860
Robust  acc:    47 /   180 =  26.111
------------------------------------
Accuracies by groups:
0, 0  acc:  9490 /  9767 =  97.164
0, 1  acc:  7518 /  7535 =  99.774
1, 0  acc:  1881 /  2480 =  75.847
1, 1  acc:    47 /   180 =  26.111
------------------------------------
Average acc: 18936 / 19962 =  94.860
Robust  acc:    47 /   180 =  26.111
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9490 /  9767 =  97.164
0, 1  acc:  7518 /  7535 =  99.774
1, 0  acc:  1881 /  2480 =  75.847
1, 1  acc:    47 /   180 =  26.111
------------------------------------
Average acc: 18936 / 19962 =  94.860
Robust  acc:    47 /   180 =  26.111
------------------------------------
Epoch:   5 | Train Loss: 0.000 | Train Acc: 94.618 | Val Loss: 0.001 | Val Acc: 94.765
Training:
Accuracies by groups:
0, 0  acc: 68794 / 71629 =  96.042
0, 1  acc: 66545 / 66874 =  99.508
1, 0  acc: 18258 / 22880 =  79.799
1, 1  acc:   413 /  1387 =  29.776
--------------------------------------
Average acc: 154010 / 162770 =  94.618
Robust  acc:   413 /  1387 =  29.776
--------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8145 /  8535 =  95.431
0, 1  acc:  8249 /  8276 =  99.674
1, 0  acc:  2391 /  2874 =  83.194
1, 1  acc:    42 /   182 =  23.077
------------------------------------
Average acc: 18827 / 19867 =  94.765
Robust  acc:    42 /   182 =  23.077
------------------------------------
Save biased model at epoch 4
replace: True
Checkpoint saved at ./model/celebA/config/stage_one_erm_model_b_epoch4_seed23.pt
-------------------------------------------
Avg Test Loss: 0.001 | Avg Test Acc: 95.301
Robust Acc: 32.778 | Best Acc: 99.695
-------------------------------------
Training, Epoch 4:
Accuracies by groups:
0, 0  acc:  9442 /  9767 =  96.672
0, 1  acc:  7512 /  7535 =  99.695
1, 0  acc:  2011 /  2480 =  81.089
1, 1  acc:    59 /   180 =  32.778
------------------------------------
Average acc: 19024 / 19962 =  95.301
Robust  acc:    59 /   180 =  32.778
------------------------------------
Accuracies by groups:
0, 0  acc:  9442 /  9767 =  96.672
0, 1  acc:  7512 /  7535 =  99.695
1, 0  acc:  2011 /  2480 =  81.089
1, 1  acc:    59 /   180 =  32.778
------------------------------------
Average acc: 19024 / 19962 =  95.301
Robust  acc:    59 /   180 =  32.778
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9442 /  9767 =  96.672
0, 1  acc:  7512 /  7535 =  99.695
1, 0  acc:  2011 /  2480 =  81.089
1, 1  acc:    59 /   180 =  32.778
------------------------------------
Average acc: 19024 / 19962 =  95.301
Robust  acc:    59 /   180 =  32.778
------------------------------------
replace: True
Checkpoint saved at ./model/celebA/config/bias-end_seed23.pt
training biased model is done
