{'rocket': 69, 'vehicle2': 19, 'veg': 4, 'mushroom': 51, 'people': 14, 'baby': 2, 'electrical_devices': 5, 'lamp': 40, 'natural_scenes': 10, 'sea': 71, '42': 42, '1': 1, '10': 10, '20': 20, '30': 30, '40': 40}
5

Class distributions (counts per class):
Retain TRAIN : [2500, 2500, 2500, 2500, 2500, 0, 2500, 2500, 2500, 2500, 2500, 2500, 2500, 2500, 2500, 2500, 2500, 2500, 2500, 2500]
Retain VALID : [500, 500, 500, 500, 500, 0, 500, 500, 500, 500, 500, 500, 500, 500, 500, 500, 500, 500, 500, 500]
Forget TRAIN : [0, 0, 0, 0, 0, 2500, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]
Forget VALID : [0, 0, 0, 0, 0, 500, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]
Epoch [0], last_lr: 0.00010, train_loss: 0.5251, val_loss: 0.7988, val_acc: 76.7914
Epoch [1], last_lr: 0.00010, train_loss: 0.3917, val_loss: 0.8203, val_acc: 77.0601
Epoch [2], last_lr: 0.00010, train_loss: 0.3232, val_loss: 0.8425, val_acc: 76.8312
Valid (Test) Dl:  10000
Train Dl:  50000
Retain Train Dl:  47500
Forget Train Dl:  2500
Retain Valid Dl:  47500
Forget Valid Dl:  2500
retain_prob Distribution: 10000 samples
test_prob Distribution: 10000 samples
forget_prob Distribution: 2500 samples
Set1 Distribution: 2500 samples
Set2 Distribution: 2500 samples
Set1 Distribution: 2500 samples
Set2 Distribution: 2500 samples
Set1 Distribution: 10000 samples
Set2 Distribution: 10000 samples
Set1 Distribution: 10000 samples
Set2 Distribution: 10000 samples
Test Accuracy: 76.83120727539062
Retain Accuracy: 98.59032440185547
Zero-Retain Forget (ZRF): 0.9888156652450562
Membership Inference Attack (MIA): 0.006
Forget vs Retain Membership Inference Attack (MIA): 0.932
Forget vs Test Membership Inference Attack (MIA): 0.899
Test vs Retain Membership Inference Attack (MIA): 0.5735
Train vs Test Membership Inference Attack (MIA): 0.5835
Forget Set Accuracy (Df): 0.078125
Method Execution Time: 1830.96 seconds
