[2021-03-20 18:07:45] Created experiment 0:
[2021-03-20 18:07:45]  - Model: vgg-pretrained
[2021-03-20 18:07:45]  - Acquisition function: lc-beam-pweighted-coreset
[2021-03-20 18:07:45] Loading cifar10 test set...
[2021-03-20 18:07:46] Experiment repeat 1/1
[2021-03-20 18:07:46] Seeded: 5
[2021-03-20 18:07:46] Using 2.00% labels of the dataset (1000/50000)
[2021-03-20 18:07:46] Randomly labelled 1000/50000
[2021-03-20 18:07:46] Showing labelled: True (1000/50000 visible, 0 redundant)
[2021-03-20 18:07:46] Seeded: 6
[2021-03-20 18:07:46] Running: experiment 0
[2021-03-20 18:07:46] Showing labelled: True (1000/50000 visible, 0 redundant)
[2021-03-20 18:07:46] Creating pretrained=True VGG16...
[2021-03-20 18:07:48] No parameter reset since we are using a pretrained model.
[2021-03-20 18:07:48] vgg-pretrained: initialized 15245130 parameters.
[2021-03-20 18:07:48] Creating trainer with model on device: cuda
[2021-03-20 18:07:53] Training vgg-pretrained across 1000 data points in cifar10...
[2021-03-20 18:08:15] Training accuracy: 0.9848
[2021-03-20 18:08:15] Testing on 10000 data points...
[2021-03-20 18:08:16] Test score for 1000 training labels: 0.7170
[2021-03-20 18:08:17] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 18:08:17] Found 50000 unlabelled features.
[2021-03-20 18:08:26] Computing distance between 1000 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 18:08:29] Searching for coresets with 1 beams...
[2021-03-20 18:08:31] Showing labelled: True (1400/50000 visible, 0 redundant)
[2021-03-20 18:08:31] Creating trainer with model on device: cuda
[2021-03-20 18:08:31] Training vgg-pretrained across 1400 data points in cifar10...
[2021-03-20 18:08:39] Training accuracy: 0.9343
[2021-03-20 18:08:39] Testing on 10000 data points...
[2021-03-20 18:08:41] Test score for 1400 training labels: 0.7442
[2021-03-20 18:08:41] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 18:08:41] Found 50000 unlabelled features.
[2021-03-20 18:08:52] Computing distance between 1400 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 18:08:56] Searching for coresets with 1 beams...
[2021-03-20 18:08:57] Showing labelled: True (1800/50000 visible, 0 redundant)
[2021-03-20 18:08:57] Creating trainer with model on device: cuda
[2021-03-20 18:08:57] Training vgg-pretrained across 1800 data points in cifar10...
[2021-03-20 18:09:08] Training accuracy: 0.9681
[2021-03-20 18:09:08] Testing on 10000 data points...
[2021-03-20 18:09:10] Test score for 1800 training labels: 0.7708
[2021-03-20 18:09:10] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 18:09:10] Found 50000 unlabelled features.
[2021-03-20 18:09:20] Computing distance between 1800 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 18:09:25] Searching for coresets with 1 beams...
[2021-03-20 18:09:27] Showing labelled: True (2200/50000 visible, 0 redundant)
[2021-03-20 18:09:27] Creating trainer with model on device: cuda
[2021-03-20 18:09:27] Training vgg-pretrained across 2200 data points in cifar10...
[2021-03-20 18:09:40] Training accuracy: 0.9757
[2021-03-20 18:09:40] Testing on 10000 data points...
[2021-03-20 18:09:42] Test score for 2200 training labels: 0.7834
[2021-03-20 18:09:42] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 18:09:42] Found 50000 unlabelled features.
[2021-03-20 18:09:53] Computing distance between 2200 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 18:09:59] Searching for coresets with 1 beams...
[2021-03-20 18:10:01] Showing labelled: True (2600/50000 visible, 0 redundant)
[2021-03-20 18:10:01] Creating trainer with model on device: cuda
[2021-03-20 18:10:01] Training vgg-pretrained across 2600 data points in cifar10...
[2021-03-20 18:10:16] Training accuracy: 0.9894
[2021-03-20 18:10:16] Testing on 10000 data points...
[2021-03-20 18:10:18] Test score for 2600 training labels: 0.7969
[2021-03-20 18:10:18] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 18:10:18] Found 50000 unlabelled features.
[2021-03-20 18:10:29] Computing distance between 2600 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 18:10:37] Searching for coresets with 1 beams...
[2021-03-20 18:10:38] Showing labelled: True (3000/50000 visible, 0 redundant)
[2021-03-20 18:10:38] Creating trainer with model on device: cuda
[2021-03-20 18:10:38] Training vgg-pretrained across 3000 data points in cifar10...
[2021-03-20 18:10:56] Training accuracy: 0.9959
[2021-03-20 18:10:56] Testing on 10000 data points...
[2021-03-20 18:10:58] Test score for 3000 training labels: 0.8028
[2021-03-20 18:10:58] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 18:10:58] Found 50000 unlabelled features.
[2021-03-20 18:11:09] Computing distance between 3000 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 18:11:17] Searching for coresets with 1 beams...
[2021-03-20 18:11:19] Showing labelled: True (3400/50000 visible, 0 redundant)
[2021-03-20 18:11:19] Creating trainer with model on device: cuda
[2021-03-20 18:11:19] Training vgg-pretrained across 3400 data points in cifar10...
[2021-03-20 18:11:39] Training accuracy: 0.9982
[2021-03-20 18:11:39] Testing on 10000 data points...
[2021-03-20 18:11:41] Test score for 3400 training labels: 0.8106
[2021-03-20 18:11:41] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 18:11:41] Found 50000 unlabelled features.
[2021-03-20 18:11:53] Computing distance between 3400 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 18:12:03] Searching for coresets with 1 beams...
[2021-03-20 18:12:04] Showing labelled: True (3800/50000 visible, 0 redundant)
[2021-03-20 18:12:04] Creating trainer with model on device: cuda
[2021-03-20 18:12:04] Training vgg-pretrained across 3800 data points in cifar10...
[2021-03-20 18:12:27] Training accuracy: 0.9993
[2021-03-20 18:12:27] Testing on 10000 data points...
[2021-03-20 18:12:29] Test score for 3800 training labels: 0.8185
[2021-03-20 18:12:29] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 18:12:29] Found 50000 unlabelled features.
[2021-03-20 18:12:39] Computing distance between 3800 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 18:12:50] Searching for coresets with 1 beams...
[2021-03-20 18:12:51] Showing labelled: True (4200/50000 visible, 0 redundant)
[2021-03-20 18:12:51] Creating trainer with model on device: cuda
[2021-03-20 18:12:51] Training vgg-pretrained across 4200 data points in cifar10...
[2021-03-20 18:13:16] Training accuracy: 0.9974
[2021-03-20 18:13:16] Testing on 10000 data points...
[2021-03-20 18:13:18] Test score for 4200 training labels: 0.8278
[2021-03-20 18:13:18] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 18:13:18] Found 50000 unlabelled features.
[2021-03-20 18:13:29] Computing distance between 4200 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 18:13:41] Searching for coresets with 1 beams...
[2021-03-20 18:13:43] Showing labelled: True (4600/50000 visible, 0 redundant)
[2021-03-20 18:13:43] Creating trainer with model on device: cuda
[2021-03-20 18:13:43] Training vgg-pretrained across 4600 data points in cifar10...
[2021-03-20 18:14:10] Training accuracy: 0.9952
[2021-03-20 18:14:10] Testing on 10000 data points...
[2021-03-20 18:14:12] Test score for 4600 training labels: 0.8328
[2021-03-20 18:14:12] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 18:14:12] Found 50000 unlabelled features.
[2021-03-20 18:14:23] Computing distance between 4600 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 18:14:36] Searching for coresets with 1 beams...
[2021-03-20 18:14:38] Showing labelled: True (5000/50000 visible, 0 redundant)
[2021-03-20 18:14:38] Creating trainer with model on device: cuda
[2021-03-20 18:14:38] Training vgg-pretrained across 5000 data points in cifar10...
[2021-03-20 18:15:08] Training accuracy: 0.9981
[2021-03-20 18:15:08] Testing on 10000 data points...
[2021-03-20 18:15:11] Test score for 5000 training labels: 0.8343
[2021-03-20 18:15:11] Updated results: ../results/cifar10/vgg_pretrained_pweighted_beam1/results.json
