text
stringlengths 5
1.13k
|
|---|
2020-02-02 11:49:00 Iteration 450 Training Loss: 1.801e-01 Loss in Target Net: 2.813e-02 |
2020-02-02 11:49:19 Iteration 500 Training Loss: 1.769e-01 Loss in Target Net: 2.658e-02 |
2020-02-02 11:49:36 Iteration 550 Training Loss: 1.777e-01 Loss in Target Net: 2.401e-02 |
2020-02-02 11:49:53 Iteration 600 Training Loss: 1.776e-01 Loss in Target Net: 2.710e-02 |
2020-02-02 11:50:10 Iteration 650 Training Loss: 1.797e-01 Loss in Target Net: 2.709e-02 |
2020-02-02 11:50:26 Iteration 700 Training Loss: 1.757e-01 Loss in Target Net: 2.756e-02 |
2020-02-02 11:50:42 Iteration 750 Training Loss: 1.754e-01 Loss in Target Net: 2.841e-02 |
2020-02-02 11:50:58 Iteration 800 Training Loss: 1.753e-01 Loss in Target Net: 2.659e-02 |
2020-02-02 11:51:17 Iteration 850 Training Loss: 1.720e-01 Loss in Target Net: 2.442e-02 |
2020-02-02 11:51:34 Iteration 900 Training Loss: 1.741e-01 Loss in Target Net: 2.831e-02 |
2020-02-02 11:51:53 Iteration 950 Training Loss: 1.717e-01 Loss in Target Net: 2.444e-02 |
2020-02-02 11:52:10 Iteration 1000 Training Loss: 1.713e-01 Loss in Target Net: 2.283e-02 |
2020-02-02 11:52:30 Iteration 1050 Training Loss: 1.752e-01 Loss in Target Net: 2.084e-02 |
2020-02-02 11:52:48 Iteration 1100 Training Loss: 1.737e-01 Loss in Target Net: 2.464e-02 |
2020-02-02 11:53:05 Iteration 1150 Training Loss: 1.706e-01 Loss in Target Net: 2.393e-02 |
2020-02-02 11:53:23 Iteration 1200 Training Loss: 1.713e-01 Loss in Target Net: 2.242e-02 |
2020-02-02 11:53:42 Iteration 1250 Training Loss: 1.731e-01 Loss in Target Net: 2.676e-02 |
2020-02-02 11:54:01 Iteration 1300 Training Loss: 1.707e-01 Loss in Target Net: 2.231e-02 |
2020-02-02 11:54:19 Iteration 1350 Training Loss: 1.673e-01 Loss in Target Net: 2.164e-02 |
2020-02-02 11:54:38 Iteration 1400 Training Loss: 1.690e-01 Loss in Target Net: 2.120e-02 |
2020-02-02 11:54:57 Iteration 1450 Training Loss: 1.733e-01 Loss in Target Net: 2.163e-02 |
2020-02-02 11:55:14 Iteration 1499 Training Loss: 1.705e-01 Loss in Target Net: 2.070e-02 |
Evaluating against victims networks |
DPN92 |
Using Adam for retraining |
Files already downloaded and verified |
2020-02-02 11:55:24, Epoch 0, Iteration 7, loss 0.404 (0.471), acc 88.462 (90.400) |
2020-02-02 11:56:22, Epoch 30, Iteration 7, loss 0.000 (0.000), acc 100.000 (100.000) |
Target Label: 6, Poison label: 8, Prediction:8, Target's Score:[-3.078151, 2.3844194, -2.3149474, -3.3393595, -1.2571094, -4.39892, 7.7854276, -2.674738, 9.481095, -2.3676348], Poisons' Predictions:[8, 8, 8, 8, 8] |
2020-02-02 11:57:22 Epoch 59, Val iteration 0, acc 93.000 (93.000) |
2020-02-02 11:57:30 Epoch 59, Val iteration 19, acc 93.000 (93.130) |
* Prec: 93.13000183105468 |
-------- |
------SUMMARY------ |
TIME ELAPSED (mins): 8 |
TARGET INDEX: 20 |
DPN92 1 |
Namespace(chk_path='chk-black-end2end', chk_subdir='poisons', device='cuda', dset_path='datasets', end2end=True, eval_poison_path='', gpu='1', lr_decay_epoch=[30, 45], mode='mean', model_resume_path='model-chks', nearest=False, net_repeat=1, num_per_class=50, original_grad=True, poison_decay_ites=[], poison_decay_ratio=0.1, poison_epsilon=0.1, poison_ites=1500, poison_label=8, poison_lr=0.04, poison_momentum=0.9, poison_num=5, poison_opt='adam', resume_poison_ite=0, retrain_bsize=64, retrain_epochs=60, retrain_lr=0.0001, retrain_momentum=0.9, retrain_opt='adam', retrain_wd=0.0005, subs_chk_name=['ckpt-%s-4800-dp0.200-droplayer0.000-seed1226.t7', 'ckpt-%s-4800-dp0.250-droplayer0.000-seed1226.t7', 'ckpt-%s-4800-dp0.300-droplayer0.000.t7'], subs_dp=[0.2, 0.25, 0.3], subset_group=0, substitute_nets=['DPN92', 'SENet18', 'ResNet50', 'ResNeXt29_2x64d'], target_index=21, target_label=6, target_net=['DPN92'], test_chk_name='ckpt-%s-4800.t7', tol=1e-06, train_data_path='datasets/CIFAR10_TRAIN_Split.pth') |
Path: chk-black-end2end/mean/1500/21 |
Selected base image indices: [213, 225, 227, 247, 249] |
2020-02-02 11:48:41 Iteration 0 Training Loss: 1.045e+00 Loss in Target Net: 1.462e+00 |
2020-02-02 11:48:58 Iteration 50 Training Loss: 2.317e-01 Loss in Target Net: 5.059e-02 |
2020-02-02 11:49:15 Iteration 100 Training Loss: 2.020e-01 Loss in Target Net: 4.343e-02 |
2020-02-02 11:49:32 Iteration 150 Training Loss: 1.897e-01 Loss in Target Net: 3.338e-02 |
2020-02-02 11:49:48 Iteration 200 Training Loss: 1.807e-01 Loss in Target Net: 3.461e-02 |
2020-02-02 11:50:05 Iteration 250 Training Loss: 1.780e-01 Loss in Target Net: 2.953e-02 |
2020-02-02 11:50:21 Iteration 300 Training Loss: 1.768e-01 Loss in Target Net: 2.764e-02 |
2020-02-02 11:50:37 Iteration 350 Training Loss: 1.693e-01 Loss in Target Net: 2.671e-02 |
2020-02-02 11:50:54 Iteration 400 Training Loss: 1.673e-01 Loss in Target Net: 2.858e-02 |
2020-02-02 11:51:12 Iteration 450 Training Loss: 1.752e-01 Loss in Target Net: 2.471e-02 |
2020-02-02 11:51:29 Iteration 500 Training Loss: 1.682e-01 Loss in Target Net: 2.593e-02 |
2020-02-02 11:51:45 Iteration 550 Training Loss: 1.660e-01 Loss in Target Net: 2.442e-02 |
2020-02-02 11:52:01 Iteration 600 Training Loss: 1.672e-01 Loss in Target Net: 2.585e-02 |
2020-02-02 11:52:18 Iteration 650 Training Loss: 1.700e-01 Loss in Target Net: 2.292e-02 |
2020-02-02 11:52:35 Iteration 700 Training Loss: 1.645e-01 Loss in Target Net: 3.039e-02 |
2020-02-02 11:52:52 Iteration 750 Training Loss: 1.651e-01 Loss in Target Net: 2.901e-02 |
2020-02-02 11:53:10 Iteration 800 Training Loss: 1.645e-01 Loss in Target Net: 2.488e-02 |
2020-02-02 11:53:27 Iteration 850 Training Loss: 1.642e-01 Loss in Target Net: 2.769e-02 |
2020-02-02 11:53:46 Iteration 900 Training Loss: 1.602e-01 Loss in Target Net: 2.798e-02 |
2020-02-02 11:54:03 Iteration 950 Training Loss: 1.662e-01 Loss in Target Net: 3.772e-02 |
2020-02-02 11:54:20 Iteration 1000 Training Loss: 1.656e-01 Loss in Target Net: 3.263e-02 |
2020-02-02 11:54:37 Iteration 1050 Training Loss: 1.627e-01 Loss in Target Net: 3.586e-02 |
2020-02-02 11:54:54 Iteration 1100 Training Loss: 1.627e-01 Loss in Target Net: 3.590e-02 |
2020-02-02 11:55:12 Iteration 1150 Training Loss: 1.627e-01 Loss in Target Net: 3.604e-02 |
2020-02-02 11:55:28 Iteration 1200 Training Loss: 1.625e-01 Loss in Target Net: 3.156e-02 |
2020-02-02 11:55:47 Iteration 1250 Training Loss: 1.584e-01 Loss in Target Net: 3.337e-02 |
2020-02-02 11:56:03 Iteration 1300 Training Loss: 1.610e-01 Loss in Target Net: 3.133e-02 |
2020-02-02 11:56:20 Iteration 1350 Training Loss: 1.607e-01 Loss in Target Net: 3.390e-02 |
2020-02-02 11:56:36 Iteration 1400 Training Loss: 1.574e-01 Loss in Target Net: 3.098e-02 |
2020-02-02 11:56:53 Iteration 1450 Training Loss: 1.573e-01 Loss in Target Net: 3.166e-02 |
2020-02-02 11:57:11 Iteration 1499 Training Loss: 1.601e-01 Loss in Target Net: 2.882e-02 |
Evaluating against victims networks |
DPN92 |
Using Adam for retraining |
Files already downloaded and verified |
2020-02-02 11:57:21, Epoch 0, Iteration 7, loss 0.332 (0.411), acc 90.385 (91.200) |
2020-02-02 11:58:19, Epoch 30, Iteration 7, loss 0.000 (0.000), acc 100.000 (100.000) |
Target Label: 6, Poison label: 8, Prediction:8, Target's Score:[-2.868253, 0.6387005, -3.5806143, -3.005225, -0.51025, -3.8126166, 7.8723106, -1.5078479, 8.921597, -1.5006611], Poisons' Predictions:[8, 8, 8, 8, 8] |
2020-02-02 11:59:19 Epoch 59, Val iteration 0, acc 93.200 (93.200) |
2020-02-02 11:59:26 Epoch 59, Val iteration 19, acc 93.000 (92.770) |
* Prec: 92.7700023651123 |
-------- |
------SUMMARY------ |
TIME ELAPSED (mins): 8 |
TARGET INDEX: 21 |
DPN92 1 |
Namespace(chk_path='chk-black-end2end', chk_subdir='poisons', device='cuda', dset_path='datasets', end2end=True, eval_poison_path='', gpu='1', lr_decay_epoch=[30, 45], mode='mean', model_resume_path='model-chks', nearest=False, net_repeat=1, num_per_class=50, original_grad=True, poison_decay_ites=[], poison_decay_ratio=0.1, poison_epsilon=0.1, poison_ites=1500, poison_label=8, poison_lr=0.04, poison_momentum=0.9, poison_num=5, poison_opt='adam', resume_poison_ite=0, retrain_bsize=64, retrain_epochs=60, retrain_lr=0.0001, retrain_momentum=0.9, retrain_opt='adam', retrain_wd=0.0005, subs_chk_name=['ckpt-%s-4800-dp0.200-droplayer0.000-seed1226.t7', 'ckpt-%s-4800-dp0.250-droplayer0.000-seed1226.t7', 'ckpt-%s-4800-dp0.300-droplayer0.000.t7'], subs_dp=[0.2, 0.25, 0.3], subset_group=0, substitute_nets=['DPN92', 'SENet18', 'ResNet50', 'ResNeXt29_2x64d'], target_index=21, target_label=6, target_net=['DPN92'], test_chk_name='ckpt-%s-4800.t7', tol=1e-06, train_data_path='datasets/CIFAR10_TRAIN_Split.pth') |
Path: chk-black-end2end/mean/1500/21 |
Selected base image indices: [213, 225, 227, 247, 249] |
2020-02-03 04:47:17 Iteration 0 Training Loss: 1.036e+00 Loss in Target Net: 1.455e+00 |
2020-02-03 04:47:59 Iteration 50 Training Loss: 2.375e-01 Loss in Target Net: 6.068e-02 |
2020-02-03 04:48:28 Iteration 100 Training Loss: 2.061e-01 Loss in Target Net: 3.840e-02 |
2020-02-03 04:49:08 Iteration 150 Training Loss: 1.899e-01 Loss in Target Net: 3.566e-02 |
2020-02-03 04:49:44 Iteration 200 Training Loss: 1.839e-01 Loss in Target Net: 4.420e-02 |
2020-02-03 04:50:29 Iteration 250 Training Loss: 1.872e-01 Loss in Target Net: 4.226e-02 |
2020-02-03 04:51:14 Iteration 300 Training Loss: 1.798e-01 Loss in Target Net: 3.682e-02 |
2020-02-03 04:51:55 Iteration 350 Training Loss: 1.765e-01 Loss in Target Net: 2.953e-02 |
2020-02-03 04:52:33 Iteration 400 Training Loss: 1.753e-01 Loss in Target Net: 3.218e-02 |
2020-02-03 04:53:19 Iteration 450 Training Loss: 1.776e-01 Loss in Target Net: 4.594e-02 |
2020-02-03 04:54:00 Iteration 500 Training Loss: 1.698e-01 Loss in Target Net: 2.976e-02 |
Subsets and Splits