[2021-03-20 02:49:14] Created experiment 0:
[2021-03-20 02:49:14]  - Model: vgg-pretrained
[2021-03-20 02:49:14]  - Acquisition function: top2entropy-beam-coreset
[2021-03-20 02:49:14] Created experiment 1:
[2021-03-20 02:49:14]  - Model: vgg-pretrained
[2021-03-20 02:49:14]  - Acquisition function: lc-beam-coreset
[2021-03-20 02:49:14] Created experiment 2:
[2021-03-20 02:49:14]  - Model: vgg-pretrained
[2021-03-20 02:49:14]  - Acquisition function: greedy-coreset
[2021-03-20 02:49:14] Created experiment 3:
[2021-03-20 02:49:14]  - Model: vgg-pretrained
[2021-03-20 02:49:14]  - Acquisition function: random
[2021-03-20 02:49:14] Loading cifar10-augmented test set...
[2021-03-20 02:49:15] Experiment repeat 1/1
[2021-03-20 02:49:15] Seeded: 5
[2021-03-20 02:49:15] Using 2.00% labels of the dataset (1000/50000)
[2021-03-20 02:49:15] Randomly labelled 1000/50000
[2021-03-20 02:49:15] Showing labelled: True (1000/50000 visible, 0 redundant)
[2021-03-20 02:49:15] Seeded: 6
[2021-03-20 02:49:15] Running: experiment 0
[2021-03-20 02:49:15] Showing labelled: True (1000/50000 visible, 0 redundant)
[2021-03-20 02:49:15] Creating pretrained=True VGG16...
[2021-03-20 02:49:17] No parameter reset since we are using a pretrained model.
[2021-03-20 02:49:17] vgg-pretrained: initialized 15245130 parameters.
[2021-03-20 02:49:17] Creating trainer with model on device: cuda
[2021-03-20 02:49:21] Training vgg-pretrained across 1000 data points in cifar10-augmented...
[2021-03-20 02:49:22] Training accuracy: 0.1083
[2021-03-20 02:49:22] Testing on 10000 data points...
[2021-03-20 02:49:24] Test score for 1000 training labels: 0.0904
[2021-03-20 02:49:24] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 02:49:24] Found 50000 unlabelled features.
[2021-03-20 02:49:36] Computing distance between 1000 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 02:49:39] Searching for coresets with 20 beams...
[2021-03-20 02:50:07] Showing labelled: True (1396/50000 visible, 4 redundant)
[2021-03-20 02:50:07] Creating pretrained=True VGG16...
[2021-03-20 02:50:09] No parameter reset since we are using a pretrained model.
[2021-03-20 02:50:09] vgg-pretrained: initialized 15245130 parameters.
[2021-03-20 02:50:09] Creating trainer with model on device: cuda
[2021-03-20 02:50:09] Training vgg-pretrained across 1400 data points in cifar10-augmented...
[2021-03-20 02:50:10] Training accuracy: 0.1004
[2021-03-20 02:50:10] Testing on 10000 data points...
[2021-03-20 02:50:12] Test score for 1400 training labels: 0.1162
[2021-03-20 02:50:12] Seeded: 6
[2021-03-20 02:50:12] Running: experiment 1
[2021-03-20 02:50:12] Showing labelled: True (1000/50000 visible, 0 redundant)
[2021-03-20 02:50:12] Creating pretrained=True VGG16...
[2021-03-20 02:50:13] No parameter reset since we are using a pretrained model.
[2021-03-20 02:50:13] vgg-pretrained: initialized 15245130 parameters.
[2021-03-20 02:50:13] Creating trainer with model on device: cuda
[2021-03-20 02:50:13] Training vgg-pretrained across 1000 data points in cifar10-augmented...
[2021-03-20 02:50:14] Training accuracy: 0.1083
[2021-03-20 02:50:14] Testing on 10000 data points...
[2021-03-20 02:50:16] Test score for 1000 training labels: 0.0904
[2021-03-20 02:50:16] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 02:50:16] Found 50000 unlabelled features.
[2021-03-20 02:50:28] Computing distance between 1000 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 02:50:31] Searching for coresets with 20 beams...
[2021-03-20 02:51:00] Showing labelled: True (1395/50000 visible, 5 redundant)
[2021-03-20 02:51:00] Creating pretrained=True VGG16...
[2021-03-20 02:51:02] No parameter reset since we are using a pretrained model.
[2021-03-20 02:51:02] vgg-pretrained: initialized 15245130 parameters.
[2021-03-20 02:51:02] Creating trainer with model on device: cuda
[2021-03-20 02:51:02] Training vgg-pretrained across 1400 data points in cifar10-augmented...
[2021-03-20 02:51:03] Training accuracy: 0.1035
[2021-03-20 02:51:03] Testing on 10000 data points...
[2021-03-20 02:51:05] Test score for 1400 training labels: 0.1162
[2021-03-20 02:51:05] Seeded: 6
[2021-03-20 02:51:05] Running: experiment 2
[2021-03-20 02:51:05] Showing labelled: True (1000/50000 visible, 0 redundant)
[2021-03-20 02:51:05] Creating pretrained=True VGG16...
[2021-03-20 02:51:07] No parameter reset since we are using a pretrained model.
[2021-03-20 02:51:07] vgg-pretrained: initialized 15245130 parameters.
[2021-03-20 02:51:07] Creating trainer with model on device: cuda
[2021-03-20 02:51:07] Training vgg-pretrained across 1000 data points in cifar10-augmented...
[2021-03-20 02:51:07] Training accuracy: 0.1083
[2021-03-20 02:51:07] Testing on 10000 data points...
[2021-03-20 02:51:09] Test score for 1000 training labels: 0.0904
[2021-03-20 02:51:09] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 02:51:09] Found 50000 unlabelled features.
[2021-03-20 02:51:22] Computing distance between 1000 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 02:51:24] Searching for coresets greedily...
[2021-03-20 02:51:25] Showing labelled: True (1395/50000 visible, 5 redundant)
[2021-03-20 02:51:25] Creating pretrained=True VGG16...
[2021-03-20 02:51:27] No parameter reset since we are using a pretrained model.
[2021-03-20 02:51:27] vgg-pretrained: initialized 15245130 parameters.
[2021-03-20 02:51:27] Creating trainer with model on device: cuda
[2021-03-20 02:51:27] Training vgg-pretrained across 1400 data points in cifar10-augmented...
[2021-03-20 02:51:28] Training accuracy: 0.1025
[2021-03-20 02:51:28] Testing on 10000 data points...
[2021-03-20 02:51:30] Test score for 1400 training labels: 0.1166
[2021-03-20 02:51:30] Seeded: 6
[2021-03-20 02:51:30] Running: experiment 3
[2021-03-20 02:51:30] Showing labelled: True (1000/50000 visible, 0 redundant)
[2021-03-20 02:51:30] Creating pretrained=True VGG16...
[2021-03-20 02:51:32] No parameter reset since we are using a pretrained model.
[2021-03-20 02:51:32] vgg-pretrained: initialized 15245130 parameters.
[2021-03-20 02:51:32] Creating trainer with model on device: cuda
[2021-03-20 02:51:32] Training vgg-pretrained across 1000 data points in cifar10-augmented...
[2021-03-20 02:51:32] Training accuracy: 0.1083
[2021-03-20 02:51:32] Testing on 10000 data points...
[2021-03-20 02:51:34] Test score for 1000 training labels: 0.0904
[2021-03-20 02:51:34] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 02:51:34] Found 50000 unlabelled features.
[2021-03-20 02:51:42] Showing labelled: True (1382/50000 visible, 18 redundant)
[2021-03-20 02:51:42] Creating pretrained=True VGG16...
[2021-03-20 02:51:44] No parameter reset since we are using a pretrained model.
[2021-03-20 02:51:44] vgg-pretrained: initialized 15245130 parameters.
[2021-03-20 02:51:44] Creating trainer with model on device: cuda
[2021-03-20 02:51:44] Training vgg-pretrained across 1400 data points in cifar10-augmented...
[2021-03-20 02:51:45] Training accuracy: 0.0897
[2021-03-20 02:51:45] Testing on 10000 data points...
[2021-03-20 02:51:47] Test score for 1400 training labels: 0.1172
[2021-03-20 02:51:47] Updated results: ../results/cifar10-aug/vgg_pretrained_top2entropy_test/results.json
