[2021-03-20 20:31:08] Created experiment 0:
[2021-03-20 20:31:08]  - Model: vgg-pretrained
[2021-03-20 20:31:08]  - Acquisition function: lc-beam-pweighted-relconf-coreset
[2021-03-20 20:31:08] Loading cifar10 test set...
[2021-03-20 20:31:09] Experiment repeat 1/1
[2021-03-20 20:31:09] Seeded: 5
[2021-03-20 20:31:09] Using 2.00% labels of the dataset (1000/50000)
[2021-03-20 20:31:09] Randomly labelled 1000/50000
[2021-03-20 20:31:09] Showing labelled: True (1000/50000 visible, 0 redundant)
[2021-03-20 20:31:09] Seeded: 6
[2021-03-20 20:31:09] Running: experiment 0
[2021-03-20 20:31:09] Showing labelled: True (1000/50000 visible, 0 redundant)
[2021-03-20 20:31:09] Creating pretrained=True VGG16...
[2021-03-20 20:31:10] No parameter reset since we are using a pretrained model.
[2021-03-20 20:31:10] vgg-pretrained: initialized 15245130 parameters.
[2021-03-20 20:31:10] Creating trainer with model on device: cuda
[2021-03-20 20:31:15] Training vgg-pretrained across 1000 data points in cifar10...
[2021-03-20 20:31:37] Training accuracy: 0.9873
[2021-03-20 20:31:37] Testing on 10000 data points...
[2021-03-20 20:31:39] Test score for 1000 training labels: 0.7167
[2021-03-20 20:31:39] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 20:31:39] Found 50000 unlabelled features.
[2021-03-20 20:31:49] Computing distance between 1000 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 20:31:52] Searching for coresets with 20 beams...
[2021-03-20 20:32:20] Showing labelled: True (1400/50000 visible, 0 redundant)
[2021-03-20 20:32:20] Creating trainer with model on device: cuda
[2021-03-20 20:32:20] Training vgg-pretrained across 1400 data points in cifar10...
[2021-03-20 20:32:28] Training accuracy: 0.9352
[2021-03-20 20:32:28] Testing on 10000 data points...
[2021-03-20 20:32:30] Test score for 1400 training labels: 0.7455
[2021-03-20 20:32:30] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 20:32:30] Found 50000 unlabelled features.
[2021-03-20 20:32:40] Computing distance between 1400 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 20:32:44] Searching for coresets with 20 beams...
[2021-03-20 20:33:12] Showing labelled: True (1800/50000 visible, 0 redundant)
[2021-03-20 20:33:12] Creating trainer with model on device: cuda
[2021-03-20 20:33:12] Training vgg-pretrained across 1800 data points in cifar10...
[2021-03-20 20:33:22] Training accuracy: 0.9687
[2021-03-20 20:33:22] Testing on 10000 data points...
[2021-03-20 20:33:24] Test score for 1800 training labels: 0.7723
[2021-03-20 20:33:24] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 20:33:24] Found 50000 unlabelled features.
[2021-03-20 20:33:34] Computing distance between 1800 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 20:33:40] Searching for coresets with 20 beams...
[2021-03-20 20:34:08] Showing labelled: True (2200/50000 visible, 0 redundant)
[2021-03-20 20:34:08] Creating trainer with model on device: cuda
[2021-03-20 20:34:08] Training vgg-pretrained across 2200 data points in cifar10...
[2021-03-20 20:34:21] Training accuracy: 0.9890
[2021-03-20 20:34:21] Testing on 10000 data points...
[2021-03-20 20:34:23] Test score for 2200 training labels: 0.7893
[2021-03-20 20:34:23] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 20:34:23] Found 50000 unlabelled features.
[2021-03-20 20:34:33] Computing distance between 2200 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 20:34:39] Searching for coresets with 20 beams...
[2021-03-20 20:35:07] Showing labelled: True (2600/50000 visible, 0 redundant)
[2021-03-20 20:35:07] Creating trainer with model on device: cuda
[2021-03-20 20:35:07] Training vgg-pretrained across 2600 data points in cifar10...
[2021-03-20 20:35:23] Training accuracy: 0.9939
[2021-03-20 20:35:23] Testing on 10000 data points...
[2021-03-20 20:35:25] Test score for 2600 training labels: 0.7972
[2021-03-20 20:35:25] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 20:35:25] Found 50000 unlabelled features.
[2021-03-20 20:35:35] Computing distance between 2600 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 20:35:42] Searching for coresets with 20 beams...
[2021-03-20 20:36:11] Showing labelled: True (3000/50000 visible, 0 redundant)
[2021-03-20 20:36:11] Creating trainer with model on device: cuda
[2021-03-20 20:36:11] Training vgg-pretrained across 3000 data points in cifar10...
[2021-03-20 20:36:29] Training accuracy: 0.9933
[2021-03-20 20:36:29] Testing on 10000 data points...
[2021-03-20 20:36:31] Test score for 3000 training labels: 0.8019
[2021-03-20 20:36:31] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 20:36:31] Found 50000 unlabelled features.
[2021-03-20 20:36:41] Computing distance between 3000 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 20:36:49] Searching for coresets with 20 beams...
[2021-03-20 20:37:18] Showing labelled: True (3400/50000 visible, 0 redundant)
[2021-03-20 20:37:18] Creating trainer with model on device: cuda
[2021-03-20 20:37:18] Training vgg-pretrained across 3400 data points in cifar10...
[2021-03-20 20:37:38] Training accuracy: 0.9974
[2021-03-20 20:37:38] Testing on 10000 data points...
[2021-03-20 20:37:40] Test score for 3400 training labels: 0.8157
[2021-03-20 20:37:40] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 20:37:40] Found 50000 unlabelled features.
[2021-03-20 20:37:50] Computing distance between 3400 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 20:38:00] Searching for coresets with 20 beams...
[2021-03-20 20:38:28] Showing labelled: True (3800/50000 visible, 0 redundant)
[2021-03-20 20:38:28] Creating trainer with model on device: cuda
[2021-03-20 20:38:28] Training vgg-pretrained across 3800 data points in cifar10...
[2021-03-20 20:38:51] Training accuracy: 0.9964
[2021-03-20 20:38:51] Testing on 10000 data points...
[2021-03-20 20:38:53] Test score for 3800 training labels: 0.8198
[2021-03-20 20:38:53] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 20:38:53] Found 50000 unlabelled features.
[2021-03-20 20:39:03] Computing distance between 3800 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 20:39:14] Searching for coresets with 20 beams...
[2021-03-20 20:39:43] Showing labelled: True (4200/50000 visible, 0 redundant)
[2021-03-20 20:39:43] Creating trainer with model on device: cuda
[2021-03-20 20:39:43] Training vgg-pretrained across 4200 data points in cifar10...
[2021-03-20 20:40:08] Training accuracy: 0.9982
[2021-03-20 20:40:08] Testing on 10000 data points...
[2021-03-20 20:40:10] Test score for 4200 training labels: 0.8278
[2021-03-20 20:40:10] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 20:40:10] Found 50000 unlabelled features.
[2021-03-20 20:40:20] Computing distance between 4200 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 20:40:32] Searching for coresets with 20 beams...
[2021-03-20 20:41:01] Showing labelled: True (4600/50000 visible, 0 redundant)
[2021-03-20 20:41:01] Creating trainer with model on device: cuda
[2021-03-20 20:41:01] Training vgg-pretrained across 4600 data points in cifar10...
[2021-03-20 20:41:28] Training accuracy: 0.9987
[2021-03-20 20:41:28] Testing on 10000 data points...
[2021-03-20 20:41:30] Test score for 4600 training labels: 0.8328
[2021-03-20 20:41:30] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 20:41:30] Found 50000 unlabelled features.
[2021-03-20 20:41:41] Computing distance between 4600 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 20:41:54] Searching for coresets with 20 beams...
[2021-03-20 20:42:22] Showing labelled: True (5000/50000 visible, 0 redundant)
[2021-03-20 20:42:22] Creating trainer with model on device: cuda
[2021-03-20 20:42:22] Training vgg-pretrained across 5000 data points in cifar10...
[2021-03-20 20:42:52] Training accuracy: 0.9981
[2021-03-20 20:42:52] Testing on 10000 data points...
[2021-03-20 20:42:54] Test score for 5000 training labels: 0.8415
[2021-03-20 20:42:54] Updated results: ../results/cifar10/vgg_pretrained_relconf/results.json
