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: 40
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=40-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: 85.086 | Val Loss: 0.003 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 71623 / 71629 =  99.992
0, 1  acc: 66870 / 66874 =  99.994
1, 0  acc:     2 / 22880 =   0.009
1, 1  acc:     0 /  1387 =   0.000
--------------------------------------
Average acc: 138495 / 162770 =  85.086
Robust  acc:     0 /  1387 =   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
------------------------------------
Save biased model at epoch 0
replace: True
Checkpoint saved at ./model/celebA/config/stage_one_erm_model_b_epoch0_seed40.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_seed40.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: 86.203 | Val Loss: 0.002 | Val Acc: 90.391
Training:
Accuracies by groups:
0, 0  acc: 71522 / 71629 =  99.851
0, 1  acc: 66874 / 66874 = 100.000
1, 0  acc:  1914 / 22880 =   8.365
1, 1  acc:     3 /  1387 =   0.216
--------------------------------------
Average acc: 140313 / 162770 =  86.203
Robust  acc:     3 /  1387 =   0.216
--------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8389 /  8535 =  98.289
0, 1  acc:  8276 /  8276 = 100.000
1, 0  acc:  1289 /  2874 =  44.850
1, 1  acc:     4 /   182 =   2.198
------------------------------------
Average acc: 17958 / 19867 =  90.391
Robust  acc:     4 /   182 =   2.198
------------------------------------
Save biased model at epoch 1
replace: True
Checkpoint saved at ./model/celebA/config/stage_one_erm_model_b_epoch1_seed40.pt
New max average-worst acc gap: 88.19329862265383
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_seed40.pt
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 91.288
Robust Acc: 2.778 | Best Acc: 100.000
-------------------------------------
Training, Epoch 1:
Accuracies by groups:
0, 0  acc:  9660 /  9767 =  98.904
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:  1023 /  2480 =  41.250
1, 1  acc:     5 /   180 =   2.778
------------------------------------
Average acc: 18223 / 19962 =  91.288
Robust  acc:     5 /   180 =   2.778
------------------------------------
Accuracies by groups:
0, 0  acc:  9660 /  9767 =  98.904
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:  1023 /  2480 =  41.250
1, 1  acc:     5 /   180 =   2.778
------------------------------------
Average acc: 18223 / 19962 =  91.288
Robust  acc:     5 /   180 =   2.778
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9660 /  9767 =  98.904
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:  1023 /  2480 =  41.250
1, 1  acc:     5 /   180 =   2.778
------------------------------------
Average acc: 18223 / 19962 =  91.288
Robust  acc:     5 /   180 =   2.778
------------------------------------
Epoch:   3 | Train Loss: 0.000 | Train Acc: 92.727 | Val Loss: 0.001 | Val Acc: 93.869
Training:
Accuracies by groups:
0, 0  acc: 69612 / 71629 =  97.184
0, 1  acc: 66771 / 66874 =  99.846
1, 0  acc: 14414 / 22880 =  62.998
1, 1  acc:   135 /  1387 =   9.733
--------------------------------------
Average acc: 150932 / 162770 =  92.727
Robust  acc:   135 /  1387 =   9.733
--------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8180 /  8535 =  95.841
0, 1  acc:  8263 /  8276 =  99.843
1, 0  acc:  2184 /  2874 =  75.992
1, 1  acc:    22 /   182 =  12.088
------------------------------------
Average acc: 18649 / 19867 =  93.869
Robust  acc:    22 /   182 =  12.088
------------------------------------
Save biased model at epoch 2
replace: True
Checkpoint saved at ./model/celebA/config/stage_one_erm_model_b_epoch2_seed40.pt
-------------------------------------------
Avg Test Loss: 0.001 | Avg Test Acc: 94.159
Robust Acc: 17.222 | Best Acc: 99.947
-------------------------------------
Training, Epoch 2:
Accuracies by groups:
0, 0  acc:  9483 /  9767 =  97.092
0, 1  acc:  7531 /  7535 =  99.947
1, 0  acc:  1751 /  2480 =  70.605
1, 1  acc:    31 /   180 =  17.222
------------------------------------
Average acc: 18796 / 19962 =  94.159
Robust  acc:    31 /   180 =  17.222
------------------------------------
Accuracies by groups:
0, 0  acc:  9483 /  9767 =  97.092
0, 1  acc:  7531 /  7535 =  99.947
1, 0  acc:  1751 /  2480 =  70.605
1, 1  acc:    31 /   180 =  17.222
------------------------------------
Average acc: 18796 / 19962 =  94.159
Robust  acc:    31 /   180 =  17.222
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9483 /  9767 =  97.092
0, 1  acc:  7531 /  7535 =  99.947
1, 0  acc:  1751 /  2480 =  70.605
1, 1  acc:    31 /   180 =  17.222
------------------------------------
Average acc: 18796 / 19962 =  94.159
Robust  acc:    31 /   180 =  17.222
------------------------------------
Epoch:   4 | Train Loss: 0.000 | Train Acc: 94.171 | Val Loss: 0.001 | Val Acc: 94.423
Training:
Accuracies by groups:
0, 0  acc: 68908 / 71629 =  96.201
0, 1  acc: 66650 / 66874 =  99.665
1, 0  acc: 17435 / 22880 =  76.202
1, 1  acc:   289 /  1387 =  20.836
--------------------------------------
Average acc: 153282 / 162770 =  94.171
Robust  acc:   289 /  1387 =  20.836
--------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8111 /  8535 =  95.032
0, 1  acc:  8256 /  8276 =  99.758
1, 0  acc:  2359 /  2874 =  82.081
1, 1  acc:    33 /   182 =  18.132
------------------------------------
Average acc: 18759 / 19867 =  94.423
Robust  acc:    33 /   182 =  18.132
------------------------------------
Save biased model at epoch 3
replace: True
Checkpoint saved at ./model/celebA/config/stage_one_erm_model_b_epoch3_seed40.pt
-------------------------------------------
Avg Test Loss: 0.001 | Avg Test Acc: 94.915
Robust Acc: 25.000 | Best Acc: 99.774
-------------------------------------
Training, Epoch 3:
Accuracies by groups:
0, 0  acc:  9428 /  9767 =  96.529
0, 1  acc:  7518 /  7535 =  99.774
1, 0  acc:  1956 /  2480 =  78.871
1, 1  acc:    45 /   180 =  25.000
------------------------------------
Average acc: 18947 / 19962 =  94.915
Robust  acc:    45 /   180 =  25.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9428 /  9767 =  96.529
0, 1  acc:  7518 /  7535 =  99.774
1, 0  acc:  1956 /  2480 =  78.871
1, 1  acc:    45 /   180 =  25.000
------------------------------------
Average acc: 18947 / 19962 =  94.915
Robust  acc:    45 /   180 =  25.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9428 /  9767 =  96.529
0, 1  acc:  7518 /  7535 =  99.774
1, 0  acc:  1956 /  2480 =  78.871
1, 1  acc:    45 /   180 =  25.000
------------------------------------
Average acc: 18947 / 19962 =  94.915
Robust  acc:    45 /   180 =  25.000
------------------------------------
Epoch:   5 | Train Loss: 0.000 | Train Acc: 94.540 | Val Loss: 0.001 | Val Acc: 94.800
Training:
Accuracies by groups:
0, 0  acc: 68818 / 71629 =  96.076
0, 1  acc: 66601 / 66874 =  99.592
1, 0  acc: 18124 / 22880 =  79.213
1, 1  acc:   340 /  1387 =  24.513
--------------------------------------
Average acc: 153883 / 162770 =  94.540
Robust  acc:   340 /  1387 =  24.513
--------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8154 /  8535 =  95.536
0, 1  acc:  8256 /  8276 =  99.758
1, 0  acc:  2386 /  2874 =  83.020
1, 1  acc:    38 /   182 =  20.879
------------------------------------
Average acc: 18834 / 19867 =  94.800
Robust  acc:    38 /   182 =  20.879
------------------------------------
Save biased model at epoch 4
replace: True
Checkpoint saved at ./model/celebA/config/stage_one_erm_model_b_epoch4_seed40.pt
-------------------------------------------
Avg Test Loss: 0.001 | Avg Test Acc: 95.276
Robust Acc: 26.667 | Best Acc: 99.774
-------------------------------------
Training, Epoch 4:
Accuracies by groups:
0, 0  acc:  9457 /  9767 =  96.826
0, 1  acc:  7518 /  7535 =  99.774
1, 0  acc:  1996 /  2480 =  80.484
1, 1  acc:    48 /   180 =  26.667
------------------------------------
Average acc: 19019 / 19962 =  95.276
Robust  acc:    48 /   180 =  26.667
------------------------------------
Accuracies by groups:
0, 0  acc:  9457 /  9767 =  96.826
0, 1  acc:  7518 /  7535 =  99.774
1, 0  acc:  1996 /  2480 =  80.484
1, 1  acc:    48 /   180 =  26.667
------------------------------------
Average acc: 19019 / 19962 =  95.276
Robust  acc:    48 /   180 =  26.667
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9457 /  9767 =  96.826
0, 1  acc:  7518 /  7535 =  99.774
1, 0  acc:  1996 /  2480 =  80.484
1, 1  acc:    48 /   180 =  26.667
------------------------------------
Average acc: 19019 / 19962 =  95.276
Robust  acc:    48 /   180 =  26.667
------------------------------------
replace: True
Checkpoint saved at ./model/celebA/config/bias-end_seed40.pt
training biased model is done
