text
stringlengths
5
1.13k
ResNet18
Using Adam for retraining
Files already downloaded and verified
2020-01-31 20:12:59, Epoch 0, Iteration 7, loss 0.877 (0.671), acc 90.385 (85.600)
2020-01-31 20:12:59, Epoch 30, Iteration 7, loss 0.018 (0.033), acc 100.000 (98.800)
Target Label: 6, Poison label: 8, Prediction:6, Target's Score:[-38.059837, -2.7982774, -17.710052, 1.30827, -33.740143, -9.269319, 10.433926, -18.773754, 9.606618, -41.93695], Poisons' Predictions:[6, 8, 8, 6, 8]
2020-01-31 20:13:00 Epoch 59, Val iteration 0, acc 93.000 (93.000)
2020-01-31 20:13:02 Epoch 59, Val iteration 19, acc 94.200 (92.910)
* Prec: 92.91000213623047
--------
DenseNet121
Using Adam for retraining
Files already downloaded and verified
2020-01-31 20:13:05, Epoch 0, Iteration 7, loss 0.245 (0.354), acc 94.231 (92.800)
2020-01-31 20:13:05, Epoch 30, Iteration 7, loss 0.002 (0.003), acc 100.000 (100.000)
Target Label: 6, Poison label: 8, Prediction:6, Target's Score:[-3.9100256, -12.739306, -12.055729, -6.397022, -0.9133542, -3.8173487, 9.006043, -26.175344, 5.8961034, -15.926775], Poisons' Predictions:[8, 8, 8, 8, 8]
2020-01-31 20:13:07 Epoch 59, Val iteration 0, acc 93.800 (93.800)
2020-01-31 20:13:11 Epoch 59, Val iteration 19, acc 92.800 (92.950)
* Prec: 92.95000267028809
--------
------SUMMARY------
TIME ELAPSED (mins): 29
TARGET INDEX: 21
DPN92 0
SENet18 0
ResNet50 1
ResNeXt29_2x64d 1
GoogLeNet 0
MobileNetV2 0
ResNet18 0
DenseNet121 0
Namespace(chk_path='chk-black-ourmean/', chk_subdir='poisons', device='cuda', dset_path='datasets', end2end=False, eval_poison_path='', gpu='2', 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=4000, 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=0.1, retrain_momentum=0.9, retrain_opt='adam', retrain_wd=0, 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', 'GoogLeNet', 'MobileNetV2'], target_index=22, target_label=6, target_net=['DPN92', 'SENet18', 'ResNet50', 'ResNeXt29_2x64d', 'GoogLeNet', 'MobileNetV2', 'ResNet18', 'DenseNet121'], test_chk_name='ckpt-%s-4800.t7', tol=1e-06, train_data_path='datasets/CIFAR10_TRAIN_Split.pth')
Path: chk-black-ourmean/mean/4000/22
Selected base image indices: [213, 225, 227, 247, 249]
2020-01-31 19:45:09 Iteration 0 Training Loss: 1.064e+00 Loss in Target Net: 4.195e-01
2020-01-31 19:45:30 Iteration 50 Training Loss: 8.687e-02 Loss in Target Net: 4.952e-03
2020-01-31 19:45:50 Iteration 100 Training Loss: 8.305e-02 Loss in Target Net: 4.519e-03
2020-01-31 19:46:11 Iteration 150 Training Loss: 7.569e-02 Loss in Target Net: 4.954e-03
2020-01-31 19:46:32 Iteration 200 Training Loss: 7.207e-02 Loss in Target Net: 7.850e-03
2020-01-31 19:46:52 Iteration 250 Training Loss: 7.705e-02 Loss in Target Net: 5.322e-03
2020-01-31 19:47:12 Iteration 300 Training Loss: 7.541e-02 Loss in Target Net: 5.092e-03
2020-01-31 19:47:32 Iteration 350 Training Loss: 7.002e-02 Loss in Target Net: 3.836e-03
2020-01-31 19:47:54 Iteration 400 Training Loss: 6.747e-02 Loss in Target Net: 6.904e-03
2020-01-31 19:48:16 Iteration 450 Training Loss: 6.745e-02 Loss in Target Net: 7.357e-03
2020-01-31 19:48:38 Iteration 500 Training Loss: 7.570e-02 Loss in Target Net: 5.638e-03
2020-01-31 19:49:00 Iteration 550 Training Loss: 7.412e-02 Loss in Target Net: 3.786e-03
2020-01-31 19:49:21 Iteration 600 Training Loss: 6.924e-02 Loss in Target Net: 4.960e-03
2020-01-31 19:49:42 Iteration 650 Training Loss: 7.007e-02 Loss in Target Net: 4.156e-03
2020-01-31 19:50:04 Iteration 700 Training Loss: 6.882e-02 Loss in Target Net: 4.071e-03
2020-01-31 19:50:27 Iteration 750 Training Loss: 6.857e-02 Loss in Target Net: 5.288e-03
2020-01-31 19:50:48 Iteration 800 Training Loss: 7.458e-02 Loss in Target Net: 4.379e-03
2020-01-31 19:51:10 Iteration 850 Training Loss: 6.938e-02 Loss in Target Net: 6.394e-03
2020-01-31 19:51:31 Iteration 900 Training Loss: 7.128e-02 Loss in Target Net: 4.644e-03
2020-01-31 19:51:54 Iteration 950 Training Loss: 6.904e-02 Loss in Target Net: 6.874e-03
2020-01-31 19:52:15 Iteration 1000 Training Loss: 6.962e-02 Loss in Target Net: 7.150e-03
2020-01-31 19:52:37 Iteration 1050 Training Loss: 6.962e-02 Loss in Target Net: 6.191e-03
2020-01-31 19:52:59 Iteration 1100 Training Loss: 6.979e-02 Loss in Target Net: 6.556e-03
2020-01-31 19:53:20 Iteration 1150 Training Loss: 7.142e-02 Loss in Target Net: 5.081e-03
2020-01-31 19:53:41 Iteration 1200 Training Loss: 6.578e-02 Loss in Target Net: 7.108e-03
2020-01-31 19:54:03 Iteration 1250 Training Loss: 6.960e-02 Loss in Target Net: 9.053e-03
2020-01-31 19:54:24 Iteration 1300 Training Loss: 7.022e-02 Loss in Target Net: 1.005e-02
2020-01-31 19:54:46 Iteration 1350 Training Loss: 6.447e-02 Loss in Target Net: 4.829e-03
2020-01-31 19:55:08 Iteration 1400 Training Loss: 7.429e-02 Loss in Target Net: 7.558e-03
2020-01-31 19:55:30 Iteration 1450 Training Loss: 6.759e-02 Loss in Target Net: 5.188e-03
2020-01-31 19:55:52 Iteration 1500 Training Loss: 6.955e-02 Loss in Target Net: 4.477e-03
2020-01-31 19:56:14 Iteration 1550 Training Loss: 6.698e-02 Loss in Target Net: 7.103e-03
2020-01-31 19:56:36 Iteration 1600 Training Loss: 7.170e-02 Loss in Target Net: 8.267e-03
2020-01-31 19:56:58 Iteration 1650 Training Loss: 6.713e-02 Loss in Target Net: 6.941e-03
2020-01-31 19:57:19 Iteration 1700 Training Loss: 7.110e-02 Loss in Target Net: 4.199e-03
2020-01-31 19:57:41 Iteration 1750 Training Loss: 6.914e-02 Loss in Target Net: 8.285e-03
2020-01-31 19:58:02 Iteration 1800 Training Loss: 6.318e-02 Loss in Target Net: 6.817e-03
2020-01-31 19:58:24 Iteration 1850 Training Loss: 6.795e-02 Loss in Target Net: 8.546e-03
2020-01-31 19:58:45 Iteration 1900 Training Loss: 6.917e-02 Loss in Target Net: 6.493e-03
2020-01-31 19:59:06 Iteration 1950 Training Loss: 6.871e-02 Loss in Target Net: 7.647e-03
2020-01-31 19:59:28 Iteration 2000 Training Loss: 7.386e-02 Loss in Target Net: 7.907e-03
2020-01-31 19:59:49 Iteration 2050 Training Loss: 6.376e-02 Loss in Target Net: 4.738e-03
2020-01-31 20:00:11 Iteration 2100 Training Loss: 6.771e-02 Loss in Target Net: 6.781e-03
2020-01-31 20:00:32 Iteration 2150 Training Loss: 7.043e-02 Loss in Target Net: 5.548e-03
2020-01-31 20:00:54 Iteration 2200 Training Loss: 6.597e-02 Loss in Target Net: 3.702e-03
2020-01-31 20:01:16 Iteration 2250 Training Loss: 6.808e-02 Loss in Target Net: 8.057e-03
2020-01-31 20:01:37 Iteration 2300 Training Loss: 6.873e-02 Loss in Target Net: 6.573e-03
2020-01-31 20:02:00 Iteration 2350 Training Loss: 6.889e-02 Loss in Target Net: 4.689e-03
2020-01-31 20:02:23 Iteration 2400 Training Loss: 6.566e-02 Loss in Target Net: 8.632e-03
2020-01-31 20:02:45 Iteration 2450 Training Loss: 6.817e-02 Loss in Target Net: 7.948e-03
2020-01-31 20:03:08 Iteration 2500 Training Loss: 7.094e-02 Loss in Target Net: 5.176e-03
2020-01-31 20:03:30 Iteration 2550 Training Loss: 6.671e-02 Loss in Target Net: 8.193e-03
2020-01-31 20:03:52 Iteration 2600 Training Loss: 6.586e-02 Loss in Target Net: 9.140e-03
2020-01-31 20:04:15 Iteration 2650 Training Loss: 7.160e-02 Loss in Target Net: 3.296e-03
2020-01-31 20:04:37 Iteration 2700 Training Loss: 7.379e-02 Loss in Target Net: 5.565e-03
2020-01-31 20:04:59 Iteration 2750 Training Loss: 7.036e-02 Loss in Target Net: 4.276e-03
2020-01-31 20:05:22 Iteration 2800 Training Loss: 6.428e-02 Loss in Target Net: 1.052e-02
2020-01-31 20:05:44 Iteration 2850 Training Loss: 6.840e-02 Loss in Target Net: 7.554e-03
2020-01-31 20:06:07 Iteration 2900 Training Loss: 7.195e-02 Loss in Target Net: 9.341e-03
2020-01-31 20:06:30 Iteration 2950 Training Loss: 7.307e-02 Loss in Target Net: 9.554e-03
2020-01-31 20:06:52 Iteration 3000 Training Loss: 6.942e-02 Loss in Target Net: 6.742e-03
2020-01-31 20:07:15 Iteration 3050 Training Loss: 6.991e-02 Loss in Target Net: 1.019e-02
2020-01-31 20:07:38 Iteration 3100 Training Loss: 6.760e-02 Loss in Target Net: 1.369e-02
2020-01-31 20:08:00 Iteration 3150 Training Loss: 7.460e-02 Loss in Target Net: 1.328e-02
2020-01-31 20:08:23 Iteration 3200 Training Loss: 7.108e-02 Loss in Target Net: 1.259e-02
2020-01-31 20:08:46 Iteration 3250 Training Loss: 6.654e-02 Loss in Target Net: 7.026e-03