[2021-03-20 02:12:48] Created experiment 0:
[2021-03-20 02:12:48]  - Model: vgg-pretrained
[2021-03-20 02:12:48]  - Acquisition function: top2entropy-beam-coreset
[2021-03-20 02:12:48] Loading cifar10 test set...
[2021-03-20 02:12:49] Experiment repeat 1/1
[2021-03-20 02:12:49] Seeded: 5
[2021-03-20 02:12:49] Using 2.00% labels of the dataset (1000/50000)
[2021-03-20 02:12:49] Randomly labelled 1000/50000
[2021-03-20 02:12:49] Showing labelled: True (1000/50000 visible, 0 redundant)
[2021-03-20 02:12:49] Seeded: 6
[2021-03-20 02:12:49] Running: experiment 0
[2021-03-20 02:12:49] Showing labelled: True (1000/50000 visible, 0 redundant)
[2021-03-20 02:12:49] Creating pretrained=True VGG16...
[2021-03-20 02:12:51] No parameter reset since we are using a pretrained model.
[2021-03-20 02:12:51] vgg-pretrained: initialized 15245130 parameters.
[2021-03-20 02:12:51] Creating trainer with model on device: cuda
[2021-03-20 02:12:56] Training vgg-pretrained across 1000 data points in cifar10...
[2021-03-20 02:13:19] Training accuracy: 0.9833
[2021-03-20 02:13:19] Testing on 10000 data points...
[2021-03-20 02:13:20] Test score for 1000 training labels: 0.7188
[2021-03-20 02:13:20] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 02:13:20] Found 50000 unlabelled features.
[2021-03-20 02:13:30] Computing distance between 1000 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 02:13:33] Searching for coresets with 20 beams...
[2021-03-20 02:14:01] Showing labelled: True (1400/50000 visible, 0 redundant)
[2021-03-20 02:14:01] Creating trainer with model on device: cuda
[2021-03-20 02:14:01] Training vgg-pretrained across 1400 data points in cifar10...
[2021-03-20 02:14:09] Training accuracy: 0.9725
[2021-03-20 02:14:09] Testing on 10000 data points...
[2021-03-20 02:14:11] Test score for 1400 training labels: 0.7440
[2021-03-20 02:14:11] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 02:14:11] Found 50000 unlabelled features.
[2021-03-20 02:14:22] Computing distance between 1400 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 02:14:26] Searching for coresets with 20 beams...
[2021-03-20 02:14:54] Showing labelled: True (1800/50000 visible, 0 redundant)
[2021-03-20 02:14:54] Creating trainer with model on device: cuda
[2021-03-20 02:14:54] Training vgg-pretrained across 1800 data points in cifar10...
[2021-03-20 02:15:05] Training accuracy: 0.9631
[2021-03-20 02:15:05] Testing on 10000 data points...
[2021-03-20 02:15:07] Test score for 1800 training labels: 0.7592
[2021-03-20 02:15:07] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 02:15:07] Found 50000 unlabelled features.
[2021-03-20 02:15:18] Computing distance between 1800 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 02:15:23] Searching for coresets with 20 beams...
[2021-03-20 02:15:51] Showing labelled: True (2200/50000 visible, 0 redundant)
[2021-03-20 02:15:51] Creating trainer with model on device: cuda
[2021-03-20 02:15:51] Training vgg-pretrained across 2200 data points in cifar10...
[2021-03-20 02:16:04] Training accuracy: 0.9889
[2021-03-20 02:16:04] Testing on 10000 data points...
[2021-03-20 02:16:07] Test score for 2200 training labels: 0.7754
[2021-03-20 02:16:07] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 02:16:07] Found 50000 unlabelled features.
[2021-03-20 02:16:17] Computing distance between 2200 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 02:16:23] Searching for coresets with 20 beams...
[2021-03-20 02:16:52] Showing labelled: True (2600/50000 visible, 0 redundant)
[2021-03-20 02:16:52] Creating trainer with model on device: cuda
[2021-03-20 02:16:52] Training vgg-pretrained across 2600 data points in cifar10...
[2021-03-20 02:17:07] Training accuracy: 0.9918
[2021-03-20 02:17:07] Testing on 10000 data points...
[2021-03-20 02:17:10] Test score for 2600 training labels: 0.7805
[2021-03-20 02:17:10] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 02:17:10] Found 50000 unlabelled features.
[2021-03-20 02:17:21] Computing distance between 2600 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 02:17:28] Searching for coresets with 20 beams...
[2021-03-20 02:17:56] Showing labelled: True (3000/50000 visible, 0 redundant)
[2021-03-20 02:17:56] Creating trainer with model on device: cuda
[2021-03-20 02:17:56] Training vgg-pretrained across 3000 data points in cifar10...
[2021-03-20 02:18:14] Training accuracy: 0.9942
[2021-03-20 02:18:14] Testing on 10000 data points...
[2021-03-20 02:18:16] Test score for 3000 training labels: 0.7935
[2021-03-20 02:18:16] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 02:18:16] Found 50000 unlabelled features.
[2021-03-20 02:18:27] Computing distance between 3000 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 02:18:35] Searching for coresets with 20 beams...
[2021-03-20 02:19:03] Showing labelled: True (3400/50000 visible, 0 redundant)
[2021-03-20 02:19:03] Creating trainer with model on device: cuda
[2021-03-20 02:19:03] Training vgg-pretrained across 3400 data points in cifar10...
[2021-03-20 02:19:24] Training accuracy: 0.9463
[2021-03-20 02:19:24] Testing on 10000 data points...
[2021-03-20 02:19:26] Test score for 3400 training labels: 0.7818
[2021-03-20 02:19:26] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 02:19:26] Found 50000 unlabelled features.
[2021-03-20 02:19:37] Computing distance between 3400 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 02:19:47] Searching for coresets with 20 beams...
[2021-03-20 02:20:16] Showing labelled: True (3800/50000 visible, 0 redundant)
[2021-03-20 02:20:16] Creating trainer with model on device: cuda
[2021-03-20 02:20:16] Training vgg-pretrained across 3800 data points in cifar10...
[2021-03-20 02:20:39] Training accuracy: 0.9944
[2021-03-20 02:20:39] Testing on 10000 data points...
[2021-03-20 02:20:41] Test score for 3800 training labels: 0.7999
[2021-03-20 02:20:41] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 02:20:41] Found 50000 unlabelled features.
[2021-03-20 02:20:52] Computing distance between 3800 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 02:21:03] Searching for coresets with 20 beams...
[2021-03-20 02:21:31] Showing labelled: True (4200/50000 visible, 0 redundant)
[2021-03-20 02:21:31] Creating trainer with model on device: cuda
[2021-03-20 02:21:31] Training vgg-pretrained across 4200 data points in cifar10...
[2021-03-20 02:21:56] Training accuracy: 0.9982
[2021-03-20 02:21:56] Testing on 10000 data points...
[2021-03-20 02:21:58] Test score for 4200 training labels: 0.8118
[2021-03-20 02:21:58] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 02:21:58] Found 50000 unlabelled features.
[2021-03-20 02:22:09] Computing distance between 4200 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 02:22:21] Searching for coresets with 20 beams...
[2021-03-20 02:22:49] Showing labelled: True (4600/50000 visible, 0 redundant)
[2021-03-20 02:22:49] Creating trainer with model on device: cuda
[2021-03-20 02:22:49] Training vgg-pretrained across 4600 data points in cifar10...
[2021-03-20 02:23:17] Training accuracy: 0.9985
[2021-03-20 02:23:17] Testing on 10000 data points...
[2021-03-20 02:23:20] Test score for 4600 training labels: 0.8185
[2021-03-20 02:23:20] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 02:23:20] Found 50000 unlabelled features.
[2021-03-20 02:23:31] Computing distance between 4600 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 02:23:44] Searching for coresets with 20 beams...
[2021-03-20 02:24:12] Showing labelled: True (5000/50000 visible, 0 redundant)
[2021-03-20 02:24:12] Creating trainer with model on device: cuda
[2021-03-20 02:24:12] Training vgg-pretrained across 5000 data points in cifar10...
[2021-03-20 02:24:42] Training accuracy: 0.9983
[2021-03-20 02:24:42] Testing on 10000 data points...
[2021-03-20 02:24:44] Test score for 5000 training labels: 0.8204
[2021-03-20 02:24:44] Updated results: ../results/cifar10/vgg_pretrained_top2entropy/results.json
