text
stringlengths 5
1.13k
|
|---|
2020-02-02 11:25:30 Iteration 50 Training Loss: 2.768e-01 Loss in Target Net: 3.452e-01 |
2020-02-02 11:25:46 Iteration 100 Training Loss: 2.462e-01 Loss in Target Net: 4.785e-01 |
2020-02-02 11:26:03 Iteration 150 Training Loss: 2.330e-01 Loss in Target Net: 4.597e-01 |
2020-02-02 11:26:19 Iteration 200 Training Loss: 2.259e-01 Loss in Target Net: 4.625e-01 |
2020-02-02 11:26:36 Iteration 250 Training Loss: 2.222e-01 Loss in Target Net: 4.020e-01 |
2020-02-02 11:26:52 Iteration 300 Training Loss: 2.224e-01 Loss in Target Net: 3.052e-01 |
2020-02-02 11:27:08 Iteration 350 Training Loss: 2.105e-01 Loss in Target Net: 3.322e-01 |
2020-02-02 11:27:25 Iteration 400 Training Loss: 2.120e-01 Loss in Target Net: 4.301e-01 |
2020-02-02 11:27:42 Iteration 450 Training Loss: 2.097e-01 Loss in Target Net: 4.030e-01 |
2020-02-02 11:27:58 Iteration 500 Training Loss: 2.094e-01 Loss in Target Net: 3.726e-01 |
2020-02-02 11:28:14 Iteration 550 Training Loss: 2.155e-01 Loss in Target Net: 3.497e-01 |
2020-02-02 11:28:30 Iteration 600 Training Loss: 2.087e-01 Loss in Target Net: 3.927e-01 |
2020-02-02 11:28:47 Iteration 650 Training Loss: 2.075e-01 Loss in Target Net: 3.537e-01 |
2020-02-02 11:29:03 Iteration 700 Training Loss: 2.082e-01 Loss in Target Net: 3.663e-01 |
2020-02-02 11:29:19 Iteration 750 Training Loss: 2.073e-01 Loss in Target Net: 2.878e-01 |
2020-02-02 11:29:36 Iteration 800 Training Loss: 2.066e-01 Loss in Target Net: 4.364e-01 |
2020-02-02 11:29:53 Iteration 850 Training Loss: 2.014e-01 Loss in Target Net: 3.691e-01 |
2020-02-02 11:30:10 Iteration 900 Training Loss: 2.037e-01 Loss in Target Net: 3.306e-01 |
2020-02-02 11:30:26 Iteration 950 Training Loss: 2.035e-01 Loss in Target Net: 3.433e-01 |
2020-02-02 11:30:43 Iteration 1000 Training Loss: 2.060e-01 Loss in Target Net: 3.904e-01 |
2020-02-02 11:31:00 Iteration 1050 Training Loss: 2.038e-01 Loss in Target Net: 3.315e-01 |
2020-02-02 11:31:16 Iteration 1100 Training Loss: 2.042e-01 Loss in Target Net: 3.285e-01 |
2020-02-02 11:31:33 Iteration 1150 Training Loss: 2.008e-01 Loss in Target Net: 4.275e-01 |
2020-02-02 11:31:49 Iteration 1200 Training Loss: 1.963e-01 Loss in Target Net: 4.001e-01 |
2020-02-02 11:32:06 Iteration 1250 Training Loss: 1.982e-01 Loss in Target Net: 3.817e-01 |
2020-02-02 11:32:23 Iteration 1300 Training Loss: 2.107e-01 Loss in Target Net: 3.856e-01 |
2020-02-02 11:32:41 Iteration 1350 Training Loss: 1.987e-01 Loss in Target Net: 3.286e-01 |
2020-02-02 11:32:57 Iteration 1400 Training Loss: 2.063e-01 Loss in Target Net: 3.952e-01 |
2020-02-02 11:33:14 Iteration 1450 Training Loss: 1.941e-01 Loss in Target Net: 3.406e-01 |
2020-02-02 11:33:30 Iteration 1499 Training Loss: 2.016e-01 Loss in Target Net: 3.851e-01 |
Evaluating against victims networks |
DPN92 |
Using Adam for retraining |
Files already downloaded and verified |
2020-02-02 11:33:40, Epoch 0, Iteration 7, loss 0.387 (0.447), acc 90.385 (90.200) |
2020-02-02 11:34:38, Epoch 30, Iteration 7, loss 0.000 (0.000), acc 100.000 (100.000) |
Target Label: 6, Poison label: 8, Prediction:5, Target's Score:[-3.9881563, -2.7717457, -2.892197, 4.1001387, -0.45415112, 6.793654, 3.187876, -3.157071, 1.5527588, -2.061669], Poisons' Predictions:[8, 8, 8, 8, 8] |
2020-02-02 11:35:38 Epoch 59, Val iteration 0, acc 92.800 (92.800) |
2020-02-02 11:35:45 Epoch 59, Val iteration 19, acc 93.800 (93.210) |
* Prec: 93.21000175476074 |
-------- |
------SUMMARY------ |
TIME ELAPSED (mins): 8 |
TARGET INDEX: 14 |
DPN92 0 |
Namespace(chk_path='chk-black-end2end', chk_subdir='poisons', device='cuda', dset_path='datasets', end2end=True, eval_poison_path='', gpu='3', 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=15, 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/1500/15 |
Selected base image indices: [213, 225, 227, 247, 249] |
2020-02-02 11:23:11 Iteration 0 Training Loss: 1.028e+00 Loss in Target Net: 1.522e+00 |
2020-02-02 11:23:27 Iteration 50 Training Loss: 2.914e-01 Loss in Target Net: 1.043e-01 |
2020-02-02 11:23:44 Iteration 100 Training Loss: 2.577e-01 Loss in Target Net: 1.044e-01 |
2020-02-02 11:23:59 Iteration 150 Training Loss: 2.446e-01 Loss in Target Net: 1.073e-01 |
2020-02-02 11:24:14 Iteration 200 Training Loss: 2.354e-01 Loss in Target Net: 8.698e-02 |
2020-02-02 11:24:32 Iteration 250 Training Loss: 2.331e-01 Loss in Target Net: 8.688e-02 |
2020-02-02 11:24:49 Iteration 300 Training Loss: 2.257e-01 Loss in Target Net: 8.483e-02 |
2020-02-02 11:25:05 Iteration 350 Training Loss: 2.194e-01 Loss in Target Net: 7.334e-02 |
2020-02-02 11:25:21 Iteration 400 Training Loss: 2.188e-01 Loss in Target Net: 7.040e-02 |
2020-02-02 11:25:36 Iteration 450 Training Loss: 2.153e-01 Loss in Target Net: 7.866e-02 |
2020-02-02 11:25:51 Iteration 500 Training Loss: 2.130e-01 Loss in Target Net: 7.713e-02 |
2020-02-02 11:26:07 Iteration 550 Training Loss: 2.151e-01 Loss in Target Net: 7.943e-02 |
2020-02-02 11:26:23 Iteration 600 Training Loss: 2.119e-01 Loss in Target Net: 8.001e-02 |
2020-02-02 11:26:40 Iteration 650 Training Loss: 2.161e-01 Loss in Target Net: 7.754e-02 |
2020-02-02 11:26:56 Iteration 700 Training Loss: 2.093e-01 Loss in Target Net: 7.471e-02 |
2020-02-02 11:27:12 Iteration 750 Training Loss: 2.115e-01 Loss in Target Net: 8.064e-02 |
2020-02-02 11:27:28 Iteration 800 Training Loss: 2.046e-01 Loss in Target Net: 8.629e-02 |
2020-02-02 11:27:44 Iteration 850 Training Loss: 2.130e-01 Loss in Target Net: 6.628e-02 |
2020-02-02 11:28:01 Iteration 900 Training Loss: 2.160e-01 Loss in Target Net: 7.420e-02 |
2020-02-02 11:28:17 Iteration 950 Training Loss: 2.072e-01 Loss in Target Net: 8.416e-02 |
2020-02-02 11:28:33 Iteration 1000 Training Loss: 2.089e-01 Loss in Target Net: 6.667e-02 |
2020-02-02 11:28:50 Iteration 1050 Training Loss: 2.033e-01 Loss in Target Net: 8.186e-02 |
2020-02-02 11:29:07 Iteration 1100 Training Loss: 2.038e-01 Loss in Target Net: 7.467e-02 |
2020-02-02 11:29:22 Iteration 1150 Training Loss: 2.058e-01 Loss in Target Net: 8.247e-02 |
2020-02-02 11:29:39 Iteration 1200 Training Loss: 2.074e-01 Loss in Target Net: 6.560e-02 |
2020-02-02 11:29:55 Iteration 1250 Training Loss: 2.057e-01 Loss in Target Net: 7.768e-02 |
2020-02-02 11:30:10 Iteration 1300 Training Loss: 2.037e-01 Loss in Target Net: 7.661e-02 |
2020-02-02 11:30:25 Iteration 1350 Training Loss: 2.017e-01 Loss in Target Net: 6.396e-02 |
2020-02-02 11:30:42 Iteration 1400 Training Loss: 2.019e-01 Loss in Target Net: 5.531e-02 |
2020-02-02 11:30:58 Iteration 1450 Training Loss: 2.059e-01 Loss in Target Net: 6.550e-02 |
2020-02-02 11:31:13 Iteration 1499 Training Loss: 2.019e-01 Loss in Target Net: 7.067e-02 |
Evaluating against victims networks |
DPN92 |
Using Adam for retraining |
Files already downloaded and verified |
2020-02-02 11:31:23, Epoch 0, Iteration 7, loss 0.316 (0.350), acc 90.385 (92.000) |
2020-02-02 11:32:20, 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:[-2.138559, -0.6141201, 0.97160745, -2.6856563, -1.1768217, -3.1684608, 3.4848573, -1.7762885, 7.5511765, -0.028830465], Poisons' Predictions:[8, 8, 8, 8, 8] |
2020-02-02 11:33:19 Epoch 59, Val iteration 0, acc 92.400 (92.400) |
2020-02-02 11:33:26 Epoch 59, Val iteration 19, acc 92.400 (92.980) |
* Prec: 92.98000106811523 |
-------- |
------SUMMARY------ |
TIME ELAPSED (mins): 8 |
TARGET INDEX: 15 |
DPN92 1 |
Namespace(chk_path='chk-black-end2end', chk_subdir='poisons', device='cuda', dset_path='datasets', end2end=True, eval_poison_path='', gpu='0', 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=16, 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/1500/16 |
Selected base image indices: [213, 225, 227, 247, 249] |
2020-02-02 11:34:42 Iteration 0 Training Loss: 1.088e+00 Loss in Target Net: 1.408e+00 |
2020-02-02 11:35:00 Iteration 50 Training Loss: 2.720e-01 Loss in Target Net: 4.415e-02 |
2020-02-02 11:35:19 Iteration 100 Training Loss: 2.388e-01 Loss in Target Net: 2.936e-02 |
Subsets and Splits