[2021-03-19 21:53:45] Created experiment 0:
[2021-03-19 21:53:45]  - Model: vgg-pretrained
[2021-03-19 21:53:45]  - Acquisition function: lc-beam-coreset
[2021-03-19 21:53:45] Loading cifar10 test set...
[2021-03-19 21:53:46] Experiment repeat 1/1
[2021-03-19 21:53:46] Seeded: 5
[2021-03-19 21:53:46] Using 2.00% labels of the dataset (1000/50000)
[2021-03-19 21:53:46] Randomly labelled 1000/50000
[2021-03-19 21:53:47] Showing labelled: True (1000/50000 visible, 0 redundant)
[2021-03-19 21:53:47] Seeded: 6
[2021-03-19 21:53:47] Running: experiment 0
[2021-03-19 21:53:47] Showing labelled: True (1000/50000 visible, 0 redundant)
[2021-03-19 21:53:47] Creating pretrained=True VGG16...
[2021-03-19 21:53:48] No parameter reset since we are using a pretrained model.
[2021-03-19 21:53:48] vgg-pretrained: initialized 15250250 parameters.
[2021-03-19 21:53:48] Creating trainer with model on device: cuda
[2021-03-19 21:53:53] Training vgg-pretrained across 1000 data points in cifar10...
[2021-03-19 21:54:02] Training accuracy: 0.9818
[2021-03-19 21:54:02] Testing on 10000 data points...
[2021-03-19 21:54:04] Test score for 1000 training labels: 0.6966
[2021-03-19 21:54:04] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-19 21:54:04] Found 50000 unlabelled features.
[2021-03-19 21:54:13] Computing distance between 1000 labelled and 50000 unlabelled vectors of length 512...
[2021-03-19 21:54:16] Searching for coresets with 20 beams...
[2021-03-19 21:54:44] Showing labelled: True (1400/50000 visible, 0 redundant)
[2021-03-19 21:54:44] Creating trainer with model on device: cuda
[2021-03-19 21:54:44] Training vgg-pretrained across 1400 data points in cifar10...
[2021-03-19 21:54:49] Training accuracy: 0.9843
[2021-03-19 21:54:49] Testing on 10000 data points...
[2021-03-19 21:54:51] Test score for 1400 training labels: 0.6329
[2021-03-19 21:54:51] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-19 21:54:51] Found 50000 unlabelled features.
[2021-03-19 21:55:01] Computing distance between 1400 labelled and 50000 unlabelled vectors of length 512...
[2021-03-19 21:55:05] Searching for coresets with 20 beams...
[2021-03-19 21:55:32] Showing labelled: True (1800/50000 visible, 0 redundant)
[2021-03-19 21:55:32] Creating trainer with model on device: cuda
[2021-03-19 21:55:32] Training vgg-pretrained across 1800 data points in cifar10...
[2021-03-19 21:55:39] Training accuracy: 0.9937
[2021-03-19 21:55:39] Testing on 10000 data points...
[2021-03-19 21:55:41] Test score for 1800 training labels: 0.7274
[2021-03-19 21:55:41] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-19 21:55:41] Found 50000 unlabelled features.
[2021-03-19 21:55:51] Computing distance between 1800 labelled and 50000 unlabelled vectors of length 512...
[2021-03-19 21:55:56] Searching for coresets with 20 beams...
[2021-03-19 21:56:23] Showing labelled: True (2200/50000 visible, 0 redundant)
[2021-03-19 21:56:23] Creating trainer with model on device: cuda
[2021-03-19 21:56:23] Training vgg-pretrained across 2200 data points in cifar10...
[2021-03-19 21:56:32] Training accuracy: 0.9965
[2021-03-19 21:56:32] Testing on 10000 data points...
[2021-03-19 21:56:34] Test score for 2200 training labels: 0.7414
[2021-03-19 21:56:34] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-19 21:56:34] Found 50000 unlabelled features.
[2021-03-19 21:56:44] Computing distance between 2200 labelled and 50000 unlabelled vectors of length 512...
[2021-03-19 21:56:50] Searching for coresets with 20 beams...
[2021-03-19 21:57:17] Showing labelled: True (2600/50000 visible, 0 redundant)
[2021-03-19 21:57:17] Creating trainer with model on device: cuda
[2021-03-19 21:57:17] Training vgg-pretrained across 2600 data points in cifar10...
[2021-03-19 21:57:27] Training accuracy: 0.9980
[2021-03-19 21:57:27] Testing on 10000 data points...
[2021-03-19 21:57:29] Test score for 2600 training labels: 0.7566
[2021-03-19 21:57:29] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-19 21:57:29] Found 50000 unlabelled features.
[2021-03-19 21:57:39] Computing distance between 2600 labelled and 50000 unlabelled vectors of length 512...
[2021-03-19 21:57:46] Searching for coresets with 20 beams...
[2021-03-19 21:58:14] Showing labelled: True (3000/50000 visible, 0 redundant)
[2021-03-19 21:58:14] Creating trainer with model on device: cuda
[2021-03-19 21:58:14] Training vgg-pretrained across 3000 data points in cifar10...
[2021-03-19 21:58:25] Training accuracy: 0.9999
[2021-03-19 21:58:25] Testing on 10000 data points...
[2021-03-19 21:58:27] Test score for 3000 training labels: 0.7570
[2021-03-19 21:58:27] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-19 21:58:27] Found 50000 unlabelled features.
[2021-03-19 21:58:38] Computing distance between 3000 labelled and 50000 unlabelled vectors of length 512...
[2021-03-19 21:58:46] Searching for coresets with 20 beams...
[2021-03-19 21:59:14] Showing labelled: True (3400/50000 visible, 0 redundant)
[2021-03-19 21:59:14] Creating trainer with model on device: cuda
[2021-03-19 21:59:14] Training vgg-pretrained across 3400 data points in cifar10...
[2021-03-19 21:59:27] Training accuracy: 0.9999
[2021-03-19 21:59:27] Testing on 10000 data points...
[2021-03-19 21:59:29] Test score for 3400 training labels: 0.7723
[2021-03-19 21:59:29] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-19 21:59:29] Found 50000 unlabelled features.
[2021-03-19 21:59:39] Computing distance between 3400 labelled and 50000 unlabelled vectors of length 512...
[2021-03-19 21:59:48] Searching for coresets with 20 beams...
[2021-03-19 22:00:16] Showing labelled: True (3800/50000 visible, 0 redundant)
[2021-03-19 22:00:16] Creating trainer with model on device: cuda
[2021-03-19 22:00:16] Training vgg-pretrained across 3800 data points in cifar10...
[2021-03-19 22:00:31] Training accuracy: 0.9999
[2021-03-19 22:00:31] Testing on 10000 data points...
[2021-03-19 22:00:33] Test score for 3800 training labels: 0.7760
[2021-03-19 22:00:33] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-19 22:00:33] Found 50000 unlabelled features.
[2021-03-19 22:00:43] Computing distance between 3800 labelled and 50000 unlabelled vectors of length 512...
[2021-03-19 22:00:53] Searching for coresets with 20 beams...
[2021-03-19 22:01:21] Showing labelled: True (4200/50000 visible, 0 redundant)
[2021-03-19 22:01:21] Creating trainer with model on device: cuda
[2021-03-19 22:01:21] Training vgg-pretrained across 4200 data points in cifar10...
[2021-03-19 22:01:37] Training accuracy: 1.0000
[2021-03-19 22:01:37] Testing on 10000 data points...
[2021-03-19 22:01:39] Test score for 4200 training labels: 0.7819
[2021-03-19 22:01:39] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-19 22:01:39] Found 50000 unlabelled features.
[2021-03-19 22:01:49] Computing distance between 4200 labelled and 50000 unlabelled vectors of length 512...
[2021-03-19 22:02:01] Searching for coresets with 20 beams...
[2021-03-19 22:02:29] Showing labelled: True (4600/50000 visible, 0 redundant)
[2021-03-19 22:02:29] Creating trainer with model on device: cuda
[2021-03-19 22:02:29] Training vgg-pretrained across 4600 data points in cifar10...
[2021-03-19 22:02:46] Training accuracy: 1.0000
[2021-03-19 22:02:46] Testing on 10000 data points...
[2021-03-19 22:02:48] Test score for 4600 training labels: 0.7825
[2021-03-19 22:02:48] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-19 22:02:48] Found 50000 unlabelled features.
[2021-03-19 22:02:59] Computing distance between 4600 labelled and 50000 unlabelled vectors of length 512...
[2021-03-19 22:03:11] Searching for coresets with 20 beams...
[2021-03-19 22:03:39] Showing labelled: True (5000/50000 visible, 0 redundant)
[2021-03-19 22:03:39] Creating trainer with model on device: cuda
[2021-03-19 22:03:39] Training vgg-pretrained across 5000 data points in cifar10...
[2021-03-19 22:03:58] Training accuracy: 1.0000
[2021-03-19 22:03:58] Testing on 10000 data points...
[2021-03-19 22:04:00] Test score for 5000 training labels: 0.7914
[2021-03-19 22:04:00] Updated results: ../results/cifar10/vgg_pretrained_lcbeam/results.json
