[2021-03-20 20:52:28] Created experiment 0:
[2021-03-20 20:52:28]  - Model: vgg-pretrained
[2021-03-20 20:52:28]  - Acquisition function: lc-beam-pweighted-relconf-optimcore
[2021-03-20 20:52:28] Loading cifar10 test set...
[2021-03-20 20:52:29] Experiment repeat 1/1
[2021-03-20 20:52:29] Seeded: 5
[2021-03-20 20:52:29] Using 2.00% labels of the dataset (1000/50000)
[2021-03-20 20:52:29] Randomly labelled 1000/50000
[2021-03-20 20:52:29] Showing labelled: True (1000/50000 visible, 0 redundant)
[2021-03-20 20:52:29] Seeded: 6
[2021-03-20 20:52:29] Running: experiment 0
[2021-03-20 20:52:29] Showing labelled: True (1000/50000 visible, 0 redundant)
[2021-03-20 20:52:29] Creating pretrained=True VGG16...
[2021-03-20 20:52:30] No parameter reset since we are using a pretrained model.
[2021-03-20 20:52:30] vgg-pretrained: initialized 15245130 parameters.
[2021-03-20 20:52:30] Creating trainer with model on device: cuda
[2021-03-20 20:52:35] Training vgg-pretrained across 1000 data points in cifar10...
[2021-03-20 20:52:58] Training accuracy: 0.9825
[2021-03-20 20:52:58] Testing on 10000 data points...
[2021-03-20 20:53:00] Test score for 1000 training labels: 0.7176
[2021-03-20 20:53:00] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 20:53:00] Found 50000 unlabelled features.
[2021-03-20 20:53:10] Computing distance between 1000 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 20:53:13] Searching for coresets with 20 beams...
[2021-03-20 20:53:41] Showing labelled: True (1400/50000 visible, 0 redundant)
[2021-03-20 20:53:41] Creating trainer with model on device: cuda
[2021-03-20 20:53:41] Training vgg-pretrained across 1400 data points in cifar10...
[2021-03-20 20:53:49] Training accuracy: 0.9423
[2021-03-20 20:53:49] Testing on 10000 data points...
[2021-03-20 20:53:51] Test score for 1400 training labels: 0.7380
[2021-03-20 20:53:51] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 20:53:51] Found 50000 unlabelled features.
[2021-03-20 20:54:02] Computing distance between 1400 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 20:54:07] Searching for coresets with 20 beams...
[2021-03-20 20:54:35] Showing labelled: True (1800/50000 visible, 0 redundant)
[2021-03-20 20:54:35] Creating trainer with model on device: cuda
[2021-03-20 20:54:35] Training vgg-pretrained across 1800 data points in cifar10...
[2021-03-20 20:54:46] Training accuracy: 0.9633
[2021-03-20 20:54:46] Testing on 10000 data points...
[2021-03-20 20:54:47] Test score for 1800 training labels: 0.7621
[2021-03-20 20:54:47] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 20:54:47] Found 50000 unlabelled features.
[2021-03-20 20:54:58] Computing distance between 1800 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 20:55:04] Searching for coresets with 20 beams...
[2021-03-20 20:55:32] Showing labelled: True (2200/50000 visible, 0 redundant)
[2021-03-20 20:55:32] Creating trainer with model on device: cuda
[2021-03-20 20:55:32] Training vgg-pretrained across 2200 data points in cifar10...
[2021-03-20 20:55:45] Training accuracy: 0.9904
[2021-03-20 20:55:45] Testing on 10000 data points...
[2021-03-20 20:55:48] Test score for 2200 training labels: 0.7851
[2021-03-20 20:55:48] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 20:55:48] Found 50000 unlabelled features.
[2021-03-20 20:55:59] Computing distance between 2200 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 20:56:05] Searching for coresets with 20 beams...
[2021-03-20 20:56:33] Showing labelled: True (2600/50000 visible, 0 redundant)
[2021-03-20 20:56:33] Creating trainer with model on device: cuda
[2021-03-20 20:56:33] Training vgg-pretrained across 2600 data points in cifar10...
[2021-03-20 20:56:49] Training accuracy: 0.9890
[2021-03-20 20:56:49] Testing on 10000 data points...
[2021-03-20 20:56:51] Test score for 2600 training labels: 0.7948
[2021-03-20 20:56:51] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 20:56:51] Found 50000 unlabelled features.
[2021-03-20 20:57:02] Computing distance between 2600 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 20:57:10] Searching for coresets with 20 beams...
[2021-03-20 20:57:39] Showing labelled: True (3000/50000 visible, 0 redundant)
[2021-03-20 20:57:39] Creating trainer with model on device: cuda
[2021-03-20 20:57:39] Training vgg-pretrained across 3000 data points in cifar10...
[2021-03-20 20:57:57] Training accuracy: 0.9942
[2021-03-20 20:57:57] Testing on 10000 data points...
[2021-03-20 20:57:59] Test score for 3000 training labels: 0.8076
[2021-03-20 20:57:59] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 20:57:59] Found 50000 unlabelled features.
[2021-03-20 20:58:10] Computing distance between 3000 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 20:58:19] Searching for coresets with 20 beams...
[2021-03-20 20:58:47] Showing labelled: True (3400/50000 visible, 0 redundant)
[2021-03-20 20:58:47] Creating trainer with model on device: cuda
[2021-03-20 20:58:47] Training vgg-pretrained across 3400 data points in cifar10...
[2021-03-20 20:59:08] Training accuracy: 0.9938
[2021-03-20 20:59:08] Testing on 10000 data points...
[2021-03-20 20:59:10] Test score for 3400 training labels: 0.8111
[2021-03-20 20:59:10] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 20:59:10] Found 50000 unlabelled features.
[2021-03-20 20:59:21] Computing distance between 3400 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 20:59:31] Searching for coresets with 20 beams...
[2021-03-20 20:59:59] Showing labelled: True (3800/50000 visible, 0 redundant)
[2021-03-20 20:59:59] Creating trainer with model on device: cuda
[2021-03-20 20:59:59] Training vgg-pretrained across 3800 data points in cifar10...
[2021-03-20 21:00:23] Training accuracy: 0.9983
[2021-03-20 21:00:23] Testing on 10000 data points...
[2021-03-20 21:00:25] Test score for 3800 training labels: 0.8186
[2021-03-20 21:00:25] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 21:00:25] Found 50000 unlabelled features.
[2021-03-20 21:00:36] Computing distance between 3800 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 21:00:47] Searching for coresets with 20 beams...
[2021-03-20 21:01:15] Showing labelled: True (4200/50000 visible, 0 redundant)
[2021-03-20 21:01:15] Creating trainer with model on device: cuda
[2021-03-20 21:01:15] Training vgg-pretrained across 4200 data points in cifar10...
[2021-03-20 21:01:41] Training accuracy: 0.9970
[2021-03-20 21:01:41] Testing on 10000 data points...
[2021-03-20 21:01:43] Test score for 4200 training labels: 0.8221
[2021-03-20 21:01:43] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 21:01:43] Found 50000 unlabelled features.
[2021-03-20 21:01:55] Computing distance between 4200 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 21:02:07] Searching for coresets with 20 beams...
[2021-03-20 21:02:35] Showing labelled: True (4600/50000 visible, 0 redundant)
[2021-03-20 21:02:35] Creating trainer with model on device: cuda
[2021-03-20 21:02:35] Training vgg-pretrained across 4600 data points in cifar10...
[2021-03-20 21:03:03] Training accuracy: 0.9987
[2021-03-20 21:03:03] Testing on 10000 data points...
[2021-03-20 21:03:05] Test score for 4600 training labels: 0.8270
[2021-03-20 21:03:05] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 21:03:05] Found 50000 unlabelled features.
[2021-03-20 21:03:16] Computing distance between 4600 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 21:03:30] Searching for coresets with 20 beams...
[2021-03-20 21:03:58] Showing labelled: True (5000/50000 visible, 0 redundant)
[2021-03-20 21:03:58] Creating trainer with model on device: cuda
[2021-03-20 21:03:58] Training vgg-pretrained across 5000 data points in cifar10...
[2021-03-20 21:04:29] Training accuracy: 0.9985
[2021-03-20 21:04:29] Testing on 10000 data points...
[2021-03-20 21:04:31] Test score for 5000 training labels: 0.8338
[2021-03-20 21:04:31] Updated results: ../results/cifar10/vgg_pretrained_optimcore/results.json
