text
stringlengths
5
1.13k
* Prec: 92.55000114440918
--------
SENet18
Using Adam for retraining
Files already downloaded and verified
2020-01-31 20:10:07, Epoch 0, Iteration 7, loss 0.328 (0.707), acc 96.154 (88.400)
2020-01-31 20:10:07, Epoch 30, Iteration 7, loss 0.223 (0.171), acc 96.154 (95.800)
Target Label: 6, Poison label: 8, Prediction:6, Target's Score:[2.7836409, 1.1651233, -10.306649, -0.8615431, 13.772531, -10.760226, 25.51193, -9.836843, 21.872595, -15.656036], Poisons' Predictions:[8, 6, 6, 6, 6]
2020-01-31 20:10:08 Epoch 59, Val iteration 0, acc 92.000 (92.000)
2020-01-31 20:10:10 Epoch 59, Val iteration 19, acc 92.600 (91.390)
* Prec: 91.39000091552734
--------
ResNet50
Using Adam for retraining
Files already downloaded and verified
2020-01-31 20:10:12, Epoch 0, Iteration 7, loss 0.406 (1.067), acc 98.077 (88.000)
2020-01-31 20:10:12, Epoch 30, Iteration 7, loss 0.001 (0.002), acc 100.000 (99.800)
Target Label: 6, Poison label: 8, Prediction:8, Target's Score:[-41.975285, -3.935853, -30.729023, -46.49646, -55.317333, -65.878395, 37.038597, -62.06877, 41.6263, -45.24795], Poisons' Predictions:[8, 8, 8, 8, 8]
2020-01-31 20:10:13 Epoch 59, Val iteration 0, acc 93.200 (93.200)
2020-01-31 20:10:18 Epoch 59, Val iteration 19, acc 93.200 (92.990)
* Prec: 92.9900016784668
--------
ResNeXt29_2x64d
Using Adam for retraining
Files already downloaded and verified
2020-01-31 20:10:20, Epoch 0, Iteration 7, loss 1.501 (2.481), acc 82.692 (72.000)
2020-01-31 20:10:20, Epoch 30, Iteration 7, loss 0.017 (0.052), acc 100.000 (98.200)
Target Label: 6, Poison label: 8, Prediction:8, Target's Score:[-30.895042, -2.022032, -3.245903, 0.60578513, -78.1211, -27.331438, 22.878893, -26.15261, 25.9752, -22.853458], Poisons' Predictions:[8, 8, 8, 8, 8]
2020-01-31 20:10:21 Epoch 59, Val iteration 0, acc 92.600 (92.600)
2020-01-31 20:10:25 Epoch 59, Val iteration 19, acc 92.800 (93.220)
* Prec: 93.22000198364258
--------
GoogLeNet
Using Adam for retraining
Files already downloaded and verified
2020-01-31 20:10:28, Epoch 0, Iteration 7, loss 0.155 (0.378), acc 92.308 (91.400)
2020-01-31 20:10:28, Epoch 30, Iteration 7, loss 0.034 (0.122), acc 100.000 (96.000)
Target Label: 6, Poison label: 8, Prediction:8, Target's Score:[-23.914228, -5.6129336, -14.531827, -0.5138366, -13.07149, -7.455347, 12.355056, -6.990944, 13.566677, -27.589523], Poisons' Predictions:[8, 8, 8, 8, 6]
2020-01-31 20:10:31 Epoch 59, Val iteration 0, acc 91.200 (91.200)
2020-01-31 20:10:35 Epoch 59, Val iteration 19, acc 93.200 (92.100)
* Prec: 92.10000114440918
--------
MobileNetV2
Using Adam for retraining
Files already downloaded and verified
2020-01-31 20:10:38, Epoch 0, Iteration 7, loss 2.006 (2.705), acc 78.846 (68.400)
2020-01-31 20:10:38, Epoch 30, Iteration 7, loss 0.299 (0.528), acc 92.308 (90.800)
Target Label: 6, Poison label: 8, Prediction:6, Target's Score:[-1.2374022, 14.394367, -12.381626, -1.069501, -16.700102, -8.715405, 35.78669, -35.543533, 23.474827, -28.434956], Poisons' Predictions:[8, 8, 8, 8, 6]
2020-01-31 20:10:39 Epoch 59, Val iteration 0, acc 89.000 (89.000)
2020-01-31 20:10:41 Epoch 59, Val iteration 19, acc 88.600 (87.140)
* Prec: 87.1400016784668
--------
ResNet18
Using Adam for retraining
Files already downloaded and verified
2020-01-31 20:10:43, Epoch 0, Iteration 7, loss 0.875 (0.692), acc 90.385 (88.400)
2020-01-31 20:10:43, Epoch 30, Iteration 7, loss 0.185 (0.089), acc 96.154 (98.200)
Target Label: 6, Poison label: 8, Prediction:6, Target's Score:[-35.713005, -9.243376, -16.65148, 0.540066, -37.750916, -8.737976, 10.298927, -23.70248, 9.9149685, -43.796925], Poisons' Predictions:[8, 8, 8, 6, 8]
2020-01-31 20:10:43 Epoch 59, Val iteration 0, acc 93.200 (93.200)
2020-01-31 20:10:45 Epoch 59, Val iteration 19, acc 93.800 (92.570)
* Prec: 92.5700023651123
--------
DenseNet121
Using Adam for retraining
Files already downloaded and verified
2020-01-31 20:10:48, Epoch 0, Iteration 7, loss 0.172 (0.401), acc 96.154 (91.400)
2020-01-31 20:10:48, Epoch 30, Iteration 7, loss 0.020 (0.020), acc 98.077 (99.200)
Target Label: 6, Poison label: 8, Prediction:6, Target's Score:[-10.918354, -12.980745, -8.677733, -3.015527, -5.090301, -6.007987, 7.7748756, -33.810375, 6.2752557, -16.10066], Poisons' Predictions:[8, 8, 8, 6, 8]
2020-01-31 20:10:50 Epoch 59, Val iteration 0, acc 93.200 (93.200)
2020-01-31 20:10:55 Epoch 59, Val iteration 19, acc 92.800 (93.200)
* Prec: 93.20000114440919
--------
------SUMMARY------
TIME ELAPSED (mins): 28
TARGET INDEX: 23
DPN92 0
SENet18 0
ResNet50 1
ResNeXt29_2x64d 1
GoogLeNet 1
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='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=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=24, 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/4000/24
Selected base image indices: [213, 225, 227, 247, 249]
2020-01-31 20:15:23 Iteration 0 Training Loss: 1.147e+00 Loss in Target Net: 4.375e-01
2020-01-31 20:15:46 Iteration 50 Training Loss: 1.039e-01 Loss in Target Net: 1.498e-02
2020-01-31 20:16:08 Iteration 100 Training Loss: 8.364e-02 Loss in Target Net: 1.404e-02
2020-01-31 20:16:31 Iteration 150 Training Loss: 7.511e-02 Loss in Target Net: 1.010e-02
2020-01-31 20:16:54 Iteration 200 Training Loss: 6.973e-02 Loss in Target Net: 8.166e-03
2020-01-31 20:17:18 Iteration 250 Training Loss: 7.099e-02 Loss in Target Net: 9.305e-03
2020-01-31 20:17:41 Iteration 300 Training Loss: 7.229e-02 Loss in Target Net: 8.337e-03
2020-01-31 20:18:04 Iteration 350 Training Loss: 7.523e-02 Loss in Target Net: 9.599e-03
2020-01-31 20:18:26 Iteration 400 Training Loss: 7.147e-02 Loss in Target Net: 6.143e-03
2020-01-31 20:18:50 Iteration 450 Training Loss: 7.045e-02 Loss in Target Net: 6.745e-03
2020-01-31 20:19:13 Iteration 500 Training Loss: 7.349e-02 Loss in Target Net: 6.599e-03
2020-01-31 20:19:36 Iteration 550 Training Loss: 7.277e-02 Loss in Target Net: 6.859e-03
2020-01-31 20:19:58 Iteration 600 Training Loss: 6.561e-02 Loss in Target Net: 1.031e-02
2020-01-31 20:20:22 Iteration 650 Training Loss: 6.716e-02 Loss in Target Net: 9.722e-03