[2021-03-20 14:46:44] Created experiment 0:
[2021-03-20 14:46:44]  - Model: vgg-pretrained
[2021-03-20 14:46:44]  - Acquisition function: least-confidence
[2021-03-20 14:46:44] Loading cifar10 test set...
[2021-03-20 14:46:45] Experiment repeat 1/1
[2021-03-20 14:46:45] Seeded: 5
[2021-03-20 14:46:45] Using 2.00% labels of the dataset (1000/50000)
[2021-03-20 14:46:45] Randomly labelled 1000/50000
[2021-03-20 14:46:45] Showing labelled: True (1000/50000 visible, 0 redundant)
[2021-03-20 14:46:45] Seeded: 6
[2021-03-20 14:46:45] Running: experiment 0
[2021-03-20 14:46:45] Showing labelled: True (1000/50000 visible, 0 redundant)
[2021-03-20 14:46:45] Creating pretrained=True VGG16...
[2021-03-20 14:46:47] No parameter reset since we are using a pretrained model.
[2021-03-20 14:46:47] vgg-pretrained: initialized 15245130 parameters.
[2021-03-20 14:46:47] Creating trainer with model on device: cuda
[2021-03-20 14:46:51] Training vgg-pretrained across 1000 data points in cifar10...
[2021-03-20 14:47:13] Training accuracy: 0.8735
[2021-03-20 14:47:13] Testing on 10000 data points...
[2021-03-20 14:47:15] Test score for 1000 training labels: 0.7145
[2021-03-20 14:47:15] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 14:47:15] Found 50000 unlabelled features.
[2021-03-20 14:47:25] Showing labelled: True (1399/50000 visible, 1 redundant)
[2021-03-20 14:47:25] Creating trainer with model on device: cuda
[2021-03-20 14:47:25] Training vgg-pretrained across 1400 data points in cifar10...
[2021-03-20 14:47:33] Training accuracy: 0.8305
[2021-03-20 14:47:33] Testing on 10000 data points...
[2021-03-20 14:47:35] Test score for 1400 training labels: 0.7388
[2021-03-20 14:47:35] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 14:47:35] Found 50000 unlabelled features.
[2021-03-20 14:47:44] Showing labelled: True (1762/50000 visible, 38 redundant)
[2021-03-20 14:47:44] Creating trainer with model on device: cuda
[2021-03-20 14:47:44] Training vgg-pretrained across 1800 data points in cifar10...
[2021-03-20 14:47:55] Training accuracy: 0.7975
[2021-03-20 14:47:55] Testing on 10000 data points...
[2021-03-20 14:47:57] Test score for 1800 training labels: 0.7327
[2021-03-20 14:47:57] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 14:47:57] Found 50000 unlabelled features.
[2021-03-20 14:48:06] Showing labelled: True (2126/50000 visible, 74 redundant)
[2021-03-20 14:48:06] Creating trainer with model on device: cuda
[2021-03-20 14:48:06] Training vgg-pretrained across 2200 data points in cifar10...
[2021-03-20 14:48:19] Training accuracy: 0.8751
[2021-03-20 14:48:19] Testing on 10000 data points...
[2021-03-20 14:48:21] Test score for 2200 training labels: 0.7710
[2021-03-20 14:48:21] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 14:48:21] Found 50000 unlabelled features.
[2021-03-20 14:48:31] Showing labelled: True (2509/50000 visible, 91 redundant)
[2021-03-20 14:48:31] Creating trainer with model on device: cuda
[2021-03-20 14:48:31] Training vgg-pretrained across 2600 data points in cifar10...
[2021-03-20 14:48:46] Training accuracy: 0.8871
[2021-03-20 14:48:46] Testing on 10000 data points...
[2021-03-20 14:48:48] Test score for 2600 training labels: 0.7812
[2021-03-20 14:48:48] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 14:48:48] Found 50000 unlabelled features.
[2021-03-20 14:48:58] Showing labelled: True (2886/50000 visible, 114 redundant)
[2021-03-20 14:48:58] Creating trainer with model on device: cuda
[2021-03-20 14:48:58] Training vgg-pretrained across 3000 data points in cifar10...
[2021-03-20 14:49:15] Training accuracy: 0.9022
[2021-03-20 14:49:15] Testing on 10000 data points...
[2021-03-20 14:49:17] Test score for 3000 training labels: 0.7929
[2021-03-20 14:49:17] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 14:49:17] Found 50000 unlabelled features.
[2021-03-20 14:49:27] Showing labelled: True (3269/50000 visible, 131 redundant)
[2021-03-20 14:49:27] Creating trainer with model on device: cuda
[2021-03-20 14:49:27] Training vgg-pretrained across 3400 data points in cifar10...
[2021-03-20 14:49:47] Training accuracy: 0.9090
[2021-03-20 14:49:47] Testing on 10000 data points...
[2021-03-20 14:49:49] Test score for 3400 training labels: 0.8004
[2021-03-20 14:49:49] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 14:49:49] Found 50000 unlabelled features.
[2021-03-20 14:49:58] Showing labelled: True (3648/50000 visible, 152 redundant)
[2021-03-20 14:49:58] Creating trainer with model on device: cuda
[2021-03-20 14:49:58] Training vgg-pretrained across 3800 data points in cifar10...
[2021-03-20 14:50:21] Training accuracy: 0.8997
[2021-03-20 14:50:21] Testing on 10000 data points...
[2021-03-20 14:50:23] Test score for 3800 training labels: 0.8114
[2021-03-20 14:50:23] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 14:50:23] Found 50000 unlabelled features.
[2021-03-20 14:50:33] Showing labelled: True (4031/50000 visible, 169 redundant)
[2021-03-20 14:50:33] Creating trainer with model on device: cuda
[2021-03-20 14:50:33] Training vgg-pretrained across 4200 data points in cifar10...
[2021-03-20 14:50:58] Training accuracy: 0.9022
[2021-03-20 14:50:58] Testing on 10000 data points...
[2021-03-20 14:51:00] Test score for 4200 training labels: 0.8171
[2021-03-20 14:51:00] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 14:51:00] Found 50000 unlabelled features.
[2021-03-20 14:51:09] Showing labelled: True (4413/50000 visible, 187 redundant)
[2021-03-20 14:51:09] Creating trainer with model on device: cuda
[2021-03-20 14:51:09] Training vgg-pretrained across 4600 data points in cifar10...
[2021-03-20 14:51:37] Training accuracy: 0.9058
[2021-03-20 14:51:37] Testing on 10000 data points...
[2021-03-20 14:51:38] Test score for 4600 training labels: 0.8269
[2021-03-20 14:51:38] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 14:51:38] Found 50000 unlabelled features.
[2021-03-20 14:51:48] Showing labelled: True (4791/50000 visible, 209 redundant)
[2021-03-20 14:51:48] Creating trainer with model on device: cuda
[2021-03-20 14:51:48] Training vgg-pretrained across 5000 data points in cifar10...
[2021-03-20 14:52:18] Training accuracy: 0.9027
[2021-03-20 14:52:18] Testing on 10000 data points...
[2021-03-20 14:52:20] Test score for 5000 training labels: 0.8308
[2021-03-20 14:52:20] Updated results: ../results/cifar10/vgg_pretrained_lc_0.1lsmooth/results.json
