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: 43
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=43-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: 84.615 | Val Loss: 0.004 | Val Acc: 84.618
Training:
Accuracies by groups:
0, 0  acc: 71227 / 71629 =  99.439
0, 1  acc: 66392 / 66874 =  99.279
1, 0  acc:   102 / 22880 =   0.446
1, 1  acc:     7 /  1387 =   0.505
--------------------------------------
Average acc: 137728 / 162770 =  84.615
Robust  acc:   102 / 22880 =   0.446
--------------------------------------
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_seed43.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_seed43.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: 0.000 | Train Acc: 85.403 | Val Loss: 0.002 | Val Acc: 87.774
Training:
Accuracies by groups:
0, 0  acc: 71612 / 71629 =  99.976
0, 1  acc: 66874 / 66874 = 100.000
1, 0  acc:   523 / 22880 =   2.286
1, 1  acc:     1 /  1387 =   0.072
--------------------------------------
Average acc: 139010 / 162770 =  85.403
Robust  acc:     1 /  1387 =   0.072
--------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8494 /  8535 =  99.520
0, 1  acc:  8276 /  8276 = 100.000
1, 0  acc:   668 /  2874 =  23.243
1, 1  acc:     0 /   182 =   0.000
------------------------------------
Average acc: 17438 / 19867 =  87.774
Robust  acc:     0 /   182 =   0.000
------------------------------------
Save biased model at epoch 1
replace: True
Checkpoint saved at ./model/celebA/config/stage_one_erm_model_b_epoch1_seed43.pt
New max average-worst acc gap: 87.77369507223032
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_seed43.pt
-------------------------------------------
Avg Test Loss: 0.002 | Avg Test Acc: 88.994
Robust Acc: 0.556 | Best Acc: 100.000
-------------------------------------
Training, Epoch 1:
Accuracies by groups:
0, 0  acc:  9740 /  9767 =  99.724
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:   489 /  2480 =  19.718
1, 1  acc:     1 /   180 =   0.556
------------------------------------
Average acc: 17765 / 19962 =  88.994
Robust  acc:     1 /   180 =   0.556
------------------------------------
Accuracies by groups:
0, 0  acc:  9740 /  9767 =  99.724
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:   489 /  2480 =  19.718
1, 1  acc:     1 /   180 =   0.556
------------------------------------
Average acc: 17765 / 19962 =  88.994
Robust  acc:     1 /   180 =   0.556
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9740 /  9767 =  99.724
0, 1  acc:  7535 /  7535 = 100.000
1, 0  acc:   489 /  2480 =  19.718
1, 1  acc:     1 /   180 =   0.556
------------------------------------
Average acc: 17765 / 19962 =  88.994
Robust  acc:     1 /   180 =   0.556
------------------------------------
Epoch:   3 | Train Loss: 0.000 | Train Acc: 91.935 | Val Loss: 0.001 | Val Acc: 93.728
Training:
Accuracies by groups:
0, 0  acc: 70004 / 71629 =  97.731
0, 1  acc: 66807 / 66874 =  99.900
1, 0  acc: 12721 / 22880 =  55.599
1, 1  acc:   110 /  1387 =   7.931
--------------------------------------
Average acc: 149642 / 162770 =  91.935
Robust  acc:   110 /  1387 =   7.931
--------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8158 /  8535 =  95.583
0, 1  acc:  8260 /  8276 =  99.807
1, 0  acc:  2182 /  2874 =  75.922
1, 1  acc:    21 /   182 =  11.538
------------------------------------
Average acc: 18621 / 19867 =  93.728
Robust  acc:    21 /   182 =  11.538
------------------------------------
Save biased model at epoch 2
replace: True
Checkpoint saved at ./model/celebA/config/stage_one_erm_model_b_epoch2_seed43.pt
-------------------------------------------
Avg Test Loss: 0.001 | Avg Test Acc: 94.039
Robust Acc: 17.778 | Best Acc: 99.907
-------------------------------------
Training, Epoch 2:
Accuracies by groups:
0, 0  acc:  9456 /  9767 =  96.816
0, 1  acc:  7528 /  7535 =  99.907
1, 0  acc:  1756 /  2480 =  70.806
1, 1  acc:    32 /   180 =  17.778
------------------------------------
Average acc: 18772 / 19962 =  94.039
Robust  acc:    32 /   180 =  17.778
------------------------------------
Accuracies by groups:
0, 0  acc:  9456 /  9767 =  96.816
0, 1  acc:  7528 /  7535 =  99.907
1, 0  acc:  1756 /  2480 =  70.806
1, 1  acc:    32 /   180 =  17.778
------------------------------------
Average acc: 18772 / 19962 =  94.039
Robust  acc:    32 /   180 =  17.778
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9456 /  9767 =  96.816
0, 1  acc:  7528 /  7535 =  99.907
1, 0  acc:  1756 /  2480 =  70.806
1, 1  acc:    32 /   180 =  17.778
------------------------------------
Average acc: 18772 / 19962 =  94.039
Robust  acc:    32 /   180 =  17.778
------------------------------------
Epoch:   4 | Train Loss: 0.000 | Train Acc: 94.044 | Val Loss: 0.001 | Val Acc: 94.363
Training:
Accuracies by groups:
0, 0  acc: 68958 / 71629 =  96.271
0, 1  acc: 66645 / 66874 =  99.658
1, 0  acc: 17188 / 22880 =  75.122
1, 1  acc:   285 /  1387 =  20.548
--------------------------------------
Average acc: 153076 / 162770 =  94.044
Robust  acc:   285 /  1387 =  20.548
--------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8124 /  8535 =  95.185
0, 1  acc:  8253 /  8276 =  99.722
1, 0  acc:  2335 /  2874 =  81.246
1, 1  acc:    35 /   182 =  19.231
------------------------------------
Average acc: 18747 / 19867 =  94.363
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_seed43.pt
-------------------------------------------
Avg Test Loss: 0.001 | Avg Test Acc: 94.825
Robust Acc: 25.000 | Best Acc: 99.774
-------------------------------------
Training, Epoch 3:
Accuracies by groups:
0, 0  acc:  9436 /  9767 =  96.611
0, 1  acc:  7518 /  7535 =  99.774
1, 0  acc:  1930 /  2480 =  77.823
1, 1  acc:    45 /   180 =  25.000
------------------------------------
Average acc: 18929 / 19962 =  94.825
Robust  acc:    45 /   180 =  25.000
------------------------------------
Accuracies by groups:
0, 0  acc:  9436 /  9767 =  96.611
0, 1  acc:  7518 /  7535 =  99.774
1, 0  acc:  1930 /  2480 =  77.823
1, 1  acc:    45 /   180 =  25.000
------------------------------------
Average acc: 18929 / 19962 =  94.825
Robust  acc:    45 /   180 =  25.000
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9436 /  9767 =  96.611
0, 1  acc:  7518 /  7535 =  99.774
1, 0  acc:  1930 /  2480 =  77.823
1, 1  acc:    45 /   180 =  25.000
------------------------------------
Average acc: 18929 / 19962 =  94.825
Robust  acc:    45 /   180 =  25.000
------------------------------------
Epoch:   5 | Train Loss: 0.000 | Train Acc: 94.566 | Val Loss: 0.001 | Val Acc: 94.649
Training:
Accuracies by groups:
0, 0  acc: 68759 / 71629 =  95.993
0, 1  acc: 66572 / 66874 =  99.548
1, 0  acc: 18215 / 22880 =  79.611
1, 1  acc:   379 /  1387 =  27.325
--------------------------------------
Average acc: 153925 / 162770 =  94.566
Robust  acc:   379 /  1387 =  27.325
--------------------------------------
Validating:
Accuracies by groups:
0, 0  acc:  8096 /  8535 =  94.856
0, 1  acc:  8244 /  8276 =  99.613
1, 0  acc:  2425 /  2874 =  84.377
1, 1  acc:    39 /   182 =  21.429
------------------------------------
Average acc: 18804 / 19867 =  94.649
Robust  acc:    39 /   182 =  21.429
------------------------------------
Save biased model at epoch 4
replace: True
Checkpoint saved at ./model/celebA/config/stage_one_erm_model_b_epoch4_seed43.pt
-------------------------------------------
Avg Test Loss: 0.001 | Avg Test Acc: 95.296
Robust Acc: 33.333 | Best Acc: 99.668
-------------------------------------
Training, Epoch 4:
Accuracies by groups:
0, 0  acc:  9414 /  9767 =  96.386
0, 1  acc:  7510 /  7535 =  99.668
1, 0  acc:  2039 /  2480 =  82.218
1, 1  acc:    60 /   180 =  33.333
------------------------------------
Average acc: 19023 / 19962 =  95.296
Robust  acc:    60 /   180 =  33.333
------------------------------------
Accuracies by groups:
0, 0  acc:  9414 /  9767 =  96.386
0, 1  acc:  7510 /  7535 =  99.668
1, 0  acc:  2039 /  2480 =  82.218
1, 1  acc:    60 /   180 =  33.333
------------------------------------
Average acc: 19023 / 19962 =  95.296
Robust  acc:    60 /   180 =  33.333
------------------------------------
Testing:
Accuracies by groups:
0, 0  acc:  9414 /  9767 =  96.386
0, 1  acc:  7510 /  7535 =  99.668
1, 0  acc:  2039 /  2480 =  82.218
1, 1  acc:    60 /   180 =  33.333
------------------------------------
Average acc: 19023 / 19962 =  95.296
Robust  acc:    60 /   180 =  33.333
------------------------------------
replace: True
Checkpoint saved at ./model/celebA/config/bias-end_seed43.pt
training biased model is done
