text
stringlengths
5
1.13k
* Prec: 93.60000114440918
--------
ResNeXt29_2x64d
Using Adam for retraining
Files already downloaded and verified
2020-01-31 21:15:28, Epoch 0, Iteration 7, loss 0.597 (2.487), acc 88.462 (67.200)
2020-01-31 21:15:29, Epoch 30, Iteration 7, loss 0.117 (0.059), acc 98.077 (98.400)
Target Label: 6, Poison label: 8, Prediction:6, Target's Score:[-25.948854, 14.625703, -3.2078762, 15.263562, -48.31069, -16.952564, 31.704498, -13.454754, 29.643555, -17.118084], Poisons' Predictions:[8, 8, 8, 8, 8]
2020-01-31 21:15:30 Epoch 59, Val iteration 0, acc 93.000 (93.000)
2020-01-31 21:15:34 Epoch 59, Val iteration 19, acc 93.600 (93.050)
* Prec: 93.05000076293945
--------
GoogLeNet
Using Adam for retraining
Files already downloaded and verified
2020-01-31 21:15:37, Epoch 0, Iteration 7, loss 0.380 (0.480), acc 92.308 (89.600)
2020-01-31 21:15:37, Epoch 30, Iteration 7, loss 0.017 (0.053), acc 98.077 (96.800)
Target Label: 6, Poison label: 8, Prediction:6, Target's Score:[-20.071428, -3.9972813, -8.344667, 1.6701244, -2.624316, -3.0010192, 10.629424, -8.710785, 9.703469, -17.731174], Poisons' Predictions:[6, 8, 8, 8, 8]
2020-01-31 21:15:39 Epoch 59, Val iteration 0, acc 91.600 (91.600)
2020-01-31 21:15:44 Epoch 59, Val iteration 19, acc 92.000 (92.010)
* Prec: 92.01000175476074
--------
MobileNetV2
Using Adam for retraining
Files already downloaded and verified
2020-01-31 21:15:46, Epoch 0, Iteration 7, loss 0.553 (3.322), acc 90.385 (70.000)
2020-01-31 21:15:46, Epoch 30, Iteration 7, loss 0.490 (0.235), acc 90.385 (94.000)
Target Label: 6, Poison label: 8, Prediction:6, Target's Score:[-16.686144, -14.513181, -14.544146, 0.98576283, -14.510549, -17.477573, 11.765283, -46.180748, 11.524577, -23.330494], Poisons' Predictions:[8, 8, 8, 8, 8]
2020-01-31 21:15:47 Epoch 59, Val iteration 0, acc 90.000 (90.000)
2020-01-31 21:15:49 Epoch 59, Val iteration 19, acc 89.600 (87.740)
* Prec: 87.7400016784668
--------
ResNet18
Using Adam for retraining
Files already downloaded and verified
2020-01-31 21:15:51, Epoch 0, Iteration 7, loss 1.027 (0.836), acc 88.462 (83.000)
2020-01-31 21:15:51, Epoch 30, Iteration 7, loss 0.022 (0.059), acc 98.077 (98.400)
Target Label: 6, Poison label: 8, Prediction:6, Target's Score:[-27.842329, -16.261932, -14.573315, -1.0095454, -40.89432, -13.023966, 9.271186, -19.910309, 6.9399757, -28.122482], Poisons' Predictions:[8, 8, 8, 8, 8]
2020-01-31 21:15:52 Epoch 59, Val iteration 0, acc 94.000 (94.000)
2020-01-31 21:15:54 Epoch 59, Val iteration 19, acc 93.000 (92.830)
* Prec: 92.83000144958496
--------
DenseNet121
Using Adam for retraining
Files already downloaded and verified
2020-01-31 21:15:57, Epoch 0, Iteration 7, loss 0.471 (0.397), acc 94.231 (92.400)
2020-01-31 21:15:57, Epoch 30, Iteration 7, loss 0.015 (0.005), acc 100.000 (100.000)
Target Label: 6, Poison label: 8, Prediction:8, Target's Score:[-11.334214, -19.07393, -12.498389, -5.3937664, -6.7309213, -8.992865, 3.93584, -33.19469, 5.7140346, -19.698147], Poisons' Predictions:[8, 8, 8, 8, 8]
2020-01-31 21:15:59 Epoch 59, Val iteration 0, acc 93.600 (93.600)
2020-01-31 21:16:03 Epoch 59, Val iteration 19, acc 93.600 (93.240)
* Prec: 93.2400016784668
--------
------SUMMARY------
TIME ELAPSED (mins): 29
TARGET INDEX: 28
DPN92 0
SENet18 0
ResNet50 0
ResNeXt29_2x64d 0
GoogLeNet 0
MobileNetV2 0
ResNet18 0
DenseNet121 1
Namespace(chk_path='chk-black-ourmean/', chk_subdir='poisons', device='cuda', dset_path='datasets', end2end=False, 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=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=29, 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/4000/29
Selected base image indices: [213, 225, 227, 247, 249]
2020-01-31 20:43:59 Iteration 0 Training Loss: 1.028e+00 Loss in Target Net: 3.510e-01
2020-01-31 20:44:22 Iteration 50 Training Loss: 1.116e-01 Loss in Target Net: 1.118e-02
2020-01-31 20:44:44 Iteration 100 Training Loss: 9.540e-02 Loss in Target Net: 7.925e-03
2020-01-31 20:45:07 Iteration 150 Training Loss: 9.262e-02 Loss in Target Net: 9.791e-03
2020-01-31 20:45:29 Iteration 200 Training Loss: 8.721e-02 Loss in Target Net: 1.002e-02
2020-01-31 20:45:49 Iteration 250 Training Loss: 8.989e-02 Loss in Target Net: 1.114e-02
2020-01-31 20:46:13 Iteration 300 Training Loss: 8.416e-02 Loss in Target Net: 1.596e-02
2020-01-31 20:46:36 Iteration 350 Training Loss: 8.484e-02 Loss in Target Net: 1.446e-02
2020-01-31 20:46:59 Iteration 400 Training Loss: 8.585e-02 Loss in Target Net: 6.726e-03
2020-01-31 20:47:21 Iteration 450 Training Loss: 8.344e-02 Loss in Target Net: 1.035e-02
2020-01-31 20:47:42 Iteration 500 Training Loss: 8.983e-02 Loss in Target Net: 1.249e-02
2020-01-31 20:48:03 Iteration 550 Training Loss: 8.414e-02 Loss in Target Net: 1.098e-02
2020-01-31 20:48:27 Iteration 600 Training Loss: 7.815e-02 Loss in Target Net: 8.082e-03
2020-01-31 20:48:49 Iteration 650 Training Loss: 8.178e-02 Loss in Target Net: 1.002e-02
2020-01-31 20:49:10 Iteration 700 Training Loss: 8.496e-02 Loss in Target Net: 9.393e-03
2020-01-31 20:49:33 Iteration 750 Training Loss: 8.304e-02 Loss in Target Net: 1.336e-02
2020-01-31 20:49:54 Iteration 800 Training Loss: 8.285e-02 Loss in Target Net: 1.109e-02
2020-01-31 20:50:16 Iteration 850 Training Loss: 8.793e-02 Loss in Target Net: 1.182e-02
2020-01-31 20:50:40 Iteration 900 Training Loss: 8.311e-02 Loss in Target Net: 9.838e-03
2020-01-31 20:51:02 Iteration 950 Training Loss: 7.214e-02 Loss in Target Net: 1.042e-02
2020-01-31 20:51:26 Iteration 1000 Training Loss: 7.652e-02 Loss in Target Net: 1.280e-02
2020-01-31 20:51:48 Iteration 1050 Training Loss: 8.171e-02 Loss in Target Net: 1.223e-02
2020-01-31 20:52:11 Iteration 1100 Training Loss: 7.439e-02 Loss in Target Net: 1.160e-02
2020-01-31 20:52:33 Iteration 1150 Training Loss: 7.757e-02 Loss in Target Net: 8.233e-03
2020-01-31 20:52:55 Iteration 1200 Training Loss: 8.278e-02 Loss in Target Net: 8.659e-03
2020-01-31 20:53:16 Iteration 1250 Training Loss: 7.474e-02 Loss in Target Net: 1.229e-02
2020-01-31 20:53:37 Iteration 1300 Training Loss: 8.648e-02 Loss in Target Net: 8.291e-03
2020-01-31 20:53:58 Iteration 1350 Training Loss: 7.916e-02 Loss in Target Net: 6.583e-03
2020-01-31 20:54:20 Iteration 1400 Training Loss: 8.457e-02 Loss in Target Net: 1.262e-02
2020-01-31 20:54:41 Iteration 1450 Training Loss: 8.532e-02 Loss in Target Net: 1.123e-02
2020-01-31 20:55:02 Iteration 1500 Training Loss: 8.164e-02 Loss in Target Net: 1.076e-02
2020-01-31 20:55:24 Iteration 1550 Training Loss: 8.227e-02 Loss in Target Net: 1.164e-02
2020-01-31 20:55:45 Iteration 1600 Training Loss: 8.180e-02 Loss in Target Net: 1.166e-02
2020-01-31 20:56:07 Iteration 1650 Training Loss: 8.090e-02 Loss in Target Net: 1.182e-02