[2021-03-20 13:40:08] Created experiment 0:
[2021-03-20 13:40:08]  - Model: vgg-pretrained
[2021-03-20 13:40:08]  - Acquisition function: least-confidence
[2021-03-20 13:40:08] Loading cifar10-augmented test set...
[2021-03-20 13:40:09] Experiment repeat 1/4
[2021-03-20 13:40:09] Seeded: 5
[2021-03-20 13:40:09] Using 2.00% labels of the dataset (1000/50000)
[2021-03-20 13:40:09] Randomly labelled 1000/50000
[2021-03-20 13:40:09] Showing labelled: True (1000/50000 visible, 0 redundant)
[2021-03-20 13:40:09] Seeded: 6
[2021-03-20 13:40:09] Running: experiment 0
[2021-03-20 13:40:09] Showing labelled: True (1000/50000 visible, 0 redundant)
[2021-03-20 13:40:09] Creating pretrained=True VGG16...
[2021-03-20 13:40:11] No parameter reset since we are using a pretrained model.
[2021-03-20 13:40:11] vgg-pretrained: initialized 15245130 parameters.
[2021-03-20 13:40:11] Creating trainer with model on device: cuda
[2021-03-20 13:40:16] Training vgg-pretrained across 1000 data points in cifar10-augmented...
[2021-03-20 13:41:19] Training accuracy: 0.9897
[2021-03-20 13:41:19] Testing on 10000 data points...
[2021-03-20 13:41:21] Test score for 1000 training labels: 0.7455
[2021-03-20 13:41:21] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 13:41:21] Found 50000 unlabelled features.
[2021-03-20 13:41:33] Showing labelled: True (1400/50000 visible, 0 redundant)
[2021-03-20 13:41:33] Creating trainer with model on device: cuda
[2021-03-20 13:41:33] Training vgg-pretrained across 1400 data points in cifar10-augmented...
[2021-03-20 13:41:54] Training accuracy: 0.9639
[2021-03-20 13:41:54] Testing on 10000 data points...
[2021-03-20 13:41:57] Test score for 1400 training labels: 0.7710
[2021-03-20 13:41:57] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 13:41:57] Found 50000 unlabelled features.
[2021-03-20 13:42:10] Showing labelled: True (1787/50000 visible, 13 redundant)
[2021-03-20 13:42:10] Creating trainer with model on device: cuda
[2021-03-20 13:42:10] Training vgg-pretrained across 1800 data points in cifar10-augmented...
[2021-03-20 13:42:38] Training accuracy: 0.9471
[2021-03-20 13:42:38] Testing on 10000 data points...
[2021-03-20 13:42:40] Test score for 1800 training labels: 0.7907
[2021-03-20 13:42:40] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 13:42:40] Found 50000 unlabelled features.
[2021-03-20 13:42:52] Showing labelled: True (2175/50000 visible, 25 redundant)
[2021-03-20 13:42:52] Creating trainer with model on device: cuda
[2021-03-20 13:42:52] Training vgg-pretrained across 2200 data points in cifar10-augmented...
[2021-03-20 13:43:27] Training accuracy: 0.9725
[2021-03-20 13:43:27] Testing on 10000 data points...
[2021-03-20 13:43:29] Test score for 2200 training labels: 0.8010
[2021-03-20 13:43:29] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 13:43:29] Found 50000 unlabelled features.
[2021-03-20 13:43:41] Showing labelled: True (2565/50000 visible, 35 redundant)
[2021-03-20 13:43:41] Creating trainer with model on device: cuda
[2021-03-20 13:43:41] Training vgg-pretrained across 2600 data points in cifar10-augmented...
[2021-03-20 13:44:22] Training accuracy: 0.9801
[2021-03-20 13:44:22] Testing on 10000 data points...
[2021-03-20 13:44:24] Test score for 2600 training labels: 0.8088
[2021-03-20 13:44:24] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 13:44:24] Found 50000 unlabelled features.
[2021-03-20 13:44:37] Showing labelled: True (2954/50000 visible, 46 redundant)
[2021-03-20 13:44:37] Creating trainer with model on device: cuda
[2021-03-20 13:44:37] Training vgg-pretrained across 3000 data points in cifar10-augmented...
[2021-03-20 13:45:25] Training accuracy: 0.9846
[2021-03-20 13:45:25] Testing on 10000 data points...
[2021-03-20 13:45:27] Test score for 3000 training labels: 0.8190
[2021-03-20 13:45:27] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 13:45:27] Found 50000 unlabelled features.
[2021-03-20 13:45:39] Showing labelled: True (3344/50000 visible, 56 redundant)
[2021-03-20 13:45:39] Creating trainer with model on device: cuda
[2021-03-20 13:45:39] Training vgg-pretrained across 3400 data points in cifar10-augmented...
[2021-03-20 13:46:33] Training accuracy: 0.9859
[2021-03-20 13:46:33] Testing on 10000 data points...
[2021-03-20 13:46:35] Test score for 3400 training labels: 0.8253
[2021-03-20 13:46:35] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 13:46:35] Found 50000 unlabelled features.
[2021-03-20 13:46:47] Showing labelled: True (3735/50000 visible, 65 redundant)
[2021-03-20 13:46:47] Creating trainer with model on device: cuda
[2021-03-20 13:46:47] Training vgg-pretrained across 3800 data points in cifar10-augmented...
[2021-03-20 13:47:49] Training accuracy: 0.9927
[2021-03-20 13:47:49] Testing on 10000 data points...
[2021-03-20 13:47:51] Test score for 3800 training labels: 0.8304
[2021-03-20 13:47:51] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 13:47:51] Found 50000 unlabelled features.
[2021-03-20 13:48:03] Showing labelled: True (4129/50000 visible, 71 redundant)
[2021-03-20 13:48:03] Creating trainer with model on device: cuda
[2021-03-20 13:48:03] Training vgg-pretrained across 4200 data points in cifar10-augmented...
[2021-03-20 13:49:10] Training accuracy: 0.9917
[2021-03-20 13:49:10] Testing on 10000 data points...
[2021-03-20 13:49:12] Test score for 4200 training labels: 0.8401
[2021-03-20 13:49:12] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 13:49:12] Found 50000 unlabelled features.
[2021-03-20 13:49:25] Showing labelled: True (4520/50000 visible, 80 redundant)
[2021-03-20 13:49:25] Creating trainer with model on device: cuda
[2021-03-20 13:49:25] Training vgg-pretrained across 4600 data points in cifar10-augmented...
[2021-03-20 13:50:38] Training accuracy: 0.9929
[2021-03-20 13:50:38] Testing on 10000 data points...
[2021-03-20 13:50:40] Test score for 4600 training labels: 0.8422
[2021-03-20 13:50:40] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 13:50:40] Found 50000 unlabelled features.
[2021-03-20 13:50:53] Showing labelled: True (4911/50000 visible, 89 redundant)
[2021-03-20 13:50:53] Creating trainer with model on device: cuda
[2021-03-20 13:50:53] Training vgg-pretrained across 5000 data points in cifar10-augmented...
[2021-03-20 13:52:13] Training accuracy: 0.9929
[2021-03-20 13:52:13] Testing on 10000 data points...
[2021-03-20 13:52:15] Test score for 5000 training labels: 0.8469
[2021-03-20 13:52:15] Experiment repeat 2/4
[2021-03-20 13:52:15] Seeded: 6
[2021-03-20 13:52:15] Using 2.00% labels of the dataset (1000/50000)
[2021-03-20 13:52:15] Randomly labelled 1000/50000
[2021-03-20 13:52:15] Showing labelled: True (1000/50000 visible, 0 redundant)
[2021-03-20 13:52:15] Seeded: 7
[2021-03-20 13:52:15] Running: experiment 0
[2021-03-20 13:52:15] Showing labelled: True (1000/50000 visible, 0 redundant)
[2021-03-20 13:52:15] Creating pretrained=True VGG16...
[2021-03-20 13:52:17] No parameter reset since we are using a pretrained model.
[2021-03-20 13:52:17] vgg-pretrained: initialized 15245130 parameters.
[2021-03-20 13:52:17] Creating trainer with model on device: cuda
[2021-03-20 13:52:17] Training vgg-pretrained across 1000 data points in cifar10-augmented...
[2021-03-20 13:53:21] Training accuracy: 0.9916
[2021-03-20 13:53:21] Testing on 10000 data points...
[2021-03-20 13:53:23] Test score for 1000 training labels: 0.7589
[2021-03-20 13:53:23] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 13:53:23] Found 50000 unlabelled features.
[2021-03-20 13:53:36] Showing labelled: True (1400/50000 visible, 0 redundant)
[2021-03-20 13:53:36] Creating trainer with model on device: cuda
[2021-03-20 13:53:36] Training vgg-pretrained across 1400 data points in cifar10-augmented...
[2021-03-20 13:53:58] Training accuracy: 0.9602
[2021-03-20 13:53:58] Testing on 10000 data points...
[2021-03-20 13:54:00] Test score for 1400 training labels: 0.7804
[2021-03-20 13:54:00] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 13:54:00] Found 50000 unlabelled features.
[2021-03-20 13:54:13] Showing labelled: True (1789/50000 visible, 11 redundant)
[2021-03-20 13:54:13] Creating trainer with model on device: cuda
[2021-03-20 13:54:13] Training vgg-pretrained across 1800 data points in cifar10-augmented...
[2021-03-20 13:55:55] Training accuracy: 0.9548
[2021-03-20 13:55:55] Testing on 10000 data points...
[2021-03-20 13:55:57] Test score for 1800 training labels: 0.7975
[2021-03-20 13:55:57] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 13:55:57] Found 50000 unlabelled features.
[2021-03-20 13:56:10] Showing labelled: True (2171/50000 visible, 29 redundant)
[2021-03-20 13:56:10] Creating trainer with model on device: cuda
[2021-03-20 13:56:10] Training vgg-pretrained across 2200 data points in cifar10-augmented...
[2021-03-20 13:56:45] Training accuracy: 0.9745
[2021-03-20 13:56:45] Testing on 10000 data points...
[2021-03-20 13:56:47] Test score for 2200 training labels: 0.8037
[2021-03-20 13:56:47] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 13:56:47] Found 50000 unlabelled features.
[2021-03-20 13:57:00] Showing labelled: True (2561/50000 visible, 39 redundant)
[2021-03-20 13:57:00] Creating trainer with model on device: cuda
[2021-03-20 13:57:00] Training vgg-pretrained across 2600 data points in cifar10-augmented...
[2021-03-20 13:57:41] Training accuracy: 0.9839
[2021-03-20 13:57:41] Testing on 10000 data points...
[2021-03-20 13:57:43] Test score for 2600 training labels: 0.8105
[2021-03-20 13:57:43] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 13:57:43] Found 50000 unlabelled features.
[2021-03-20 13:57:55] Showing labelled: True (2952/50000 visible, 48 redundant)
[2021-03-20 13:57:55] Creating trainer with model on device: cuda
[2021-03-20 13:57:55] Training vgg-pretrained across 3000 data points in cifar10-augmented...
[2021-03-20 13:58:42] Training accuracy: 0.9900
[2021-03-20 13:58:42] Testing on 10000 data points...
[2021-03-20 13:58:44] Test score for 3000 training labels: 0.8198
[2021-03-20 13:58:44] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 13:58:44] Found 50000 unlabelled features.
[2021-03-20 13:58:57] Showing labelled: True (3341/50000 visible, 59 redundant)
[2021-03-20 13:58:57] Creating trainer with model on device: cuda
[2021-03-20 13:58:57] Training vgg-pretrained across 3400 data points in cifar10-augmented...
[2021-03-20 13:59:52] Training accuracy: 0.9859
[2021-03-20 13:59:52] Testing on 10000 data points...
[2021-03-20 13:59:53] Test score for 3400 training labels: 0.8298
[2021-03-20 13:59:53] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 13:59:53] Found 50000 unlabelled features.
[2021-03-20 14:00:07] Showing labelled: True (3730/50000 visible, 70 redundant)
[2021-03-20 14:00:07] Creating trainer with model on device: cuda
[2021-03-20 14:00:07] Training vgg-pretrained across 3800 data points in cifar10-augmented...
[2021-03-20 14:01:09] Training accuracy: 0.9897
[2021-03-20 14:01:09] Testing on 10000 data points...
[2021-03-20 14:01:11] Test score for 3800 training labels: 0.8331
[2021-03-20 14:01:11] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 14:01:11] Found 50000 unlabelled features.
[2021-03-20 14:01:24] Showing labelled: True (4123/50000 visible, 77 redundant)
[2021-03-20 14:01:24] Creating trainer with model on device: cuda
[2021-03-20 14:01:24] Training vgg-pretrained across 4200 data points in cifar10-augmented...
[2021-03-20 14:02:31] Training accuracy: 0.9937
[2021-03-20 14:02:31] Testing on 10000 data points...
[2021-03-20 14:02:32] Test score for 4200 training labels: 0.8441
[2021-03-20 14:02:32] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 14:02:32] Found 50000 unlabelled features.
[2021-03-20 14:02:45] Showing labelled: True (4510/50000 visible, 90 redundant)
[2021-03-20 14:02:45] Creating trainer with model on device: cuda
[2021-03-20 14:02:45] Training vgg-pretrained across 4600 data points in cifar10-augmented...
[2021-03-20 14:03:57] Training accuracy: 0.9925
[2021-03-20 14:03:57] Testing on 10000 data points...
[2021-03-20 14:03:59] Test score for 4600 training labels: 0.8477
[2021-03-20 14:04:00] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 14:04:00] Found 50000 unlabelled features.
[2021-03-20 14:04:13] Showing labelled: True (4900/50000 visible, 100 redundant)
[2021-03-20 14:04:13] Creating trainer with model on device: cuda
[2021-03-20 14:04:13] Training vgg-pretrained across 5000 data points in cifar10-augmented...
[2021-03-20 14:05:31] Training accuracy: 0.9945
[2021-03-20 14:05:31] Testing on 10000 data points...
[2021-03-20 14:05:33] Test score for 5000 training labels: 0.8504
[2021-03-20 14:05:33] Experiment repeat 3/4
[2021-03-20 14:05:33] Seeded: 7
[2021-03-20 14:05:33] Using 2.00% labels of the dataset (1000/50000)
[2021-03-20 14:05:33] Randomly labelled 1000/50000
[2021-03-20 14:05:33] Showing labelled: True (1000/50000 visible, 0 redundant)
[2021-03-20 14:05:33] Seeded: 8
[2021-03-20 14:05:33] Running: experiment 0
[2021-03-20 14:05:33] Showing labelled: True (1000/50000 visible, 0 redundant)
[2021-03-20 14:05:33] Creating pretrained=True VGG16...
[2021-03-20 14:05:35] No parameter reset since we are using a pretrained model.
[2021-03-20 14:05:35] vgg-pretrained: initialized 15245130 parameters.
[2021-03-20 14:05:35] Creating trainer with model on device: cuda
[2021-03-20 14:05:35] Training vgg-pretrained across 1000 data points in cifar10-augmented...
[2021-03-20 14:06:38] Training accuracy: 0.9890
[2021-03-20 14:06:38] Testing on 10000 data points...
[2021-03-20 14:06:40] Test score for 1000 training labels: 0.7440
[2021-03-20 14:06:40] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 14:06:40] Found 50000 unlabelled features.
[2021-03-20 14:06:53] Showing labelled: True (1399/50000 visible, 1 redundant)
[2021-03-20 14:06:53] Creating trainer with model on device: cuda
[2021-03-20 14:06:53] Training vgg-pretrained across 1400 data points in cifar10-augmented...
[2021-03-20 14:07:15] Training accuracy: 0.9539
[2021-03-20 14:07:15] Testing on 10000 data points...
[2021-03-20 14:07:17] Test score for 1400 training labels: 0.7621
[2021-03-20 14:07:17] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 14:07:17] Found 50000 unlabelled features.
[2021-03-20 14:07:30] Showing labelled: True (1784/50000 visible, 16 redundant)
[2021-03-20 14:07:30] Creating trainer with model on device: cuda
[2021-03-20 14:07:30] Training vgg-pretrained across 1800 data points in cifar10-augmented...
[2021-03-20 14:07:58] Training accuracy: 0.9625
[2021-03-20 14:07:58] Testing on 10000 data points...
[2021-03-20 14:08:00] Test score for 1800 training labels: 0.7848
[2021-03-20 14:08:00] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 14:08:00] Found 50000 unlabelled features.
[2021-03-20 14:08:13] Showing labelled: True (2172/50000 visible, 28 redundant)
[2021-03-20 14:08:13] Creating trainer with model on device: cuda
[2021-03-20 14:08:13] Training vgg-pretrained across 2200 data points in cifar10-augmented...
[2021-03-20 14:08:49] Training accuracy: 0.9738
[2021-03-20 14:08:49] Testing on 10000 data points...
[2021-03-20 14:08:51] Test score for 2200 training labels: 0.7944
[2021-03-20 14:08:51] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 14:08:51] Found 50000 unlabelled features.
[2021-03-20 14:09:03] Showing labelled: True (2554/50000 visible, 46 redundant)
[2021-03-20 14:09:03] Creating trainer with model on device: cuda
[2021-03-20 14:09:03] Training vgg-pretrained across 2600 data points in cifar10-augmented...
[2021-03-20 14:09:44] Training accuracy: 0.9758
[2021-03-20 14:09:44] Testing on 10000 data points...
[2021-03-20 14:09:46] Test score for 2600 training labels: 0.8048
[2021-03-20 14:09:46] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 14:09:46] Found 50000 unlabelled features.
[2021-03-20 14:09:59] Showing labelled: True (2945/50000 visible, 55 redundant)
[2021-03-20 14:09:59] Creating trainer with model on device: cuda
[2021-03-20 14:09:59] Training vgg-pretrained across 3000 data points in cifar10-augmented...
[2021-03-20 14:10:46] Training accuracy: 0.9877
[2021-03-20 14:10:46] Testing on 10000 data points...
[2021-03-20 14:10:48] Test score for 3000 training labels: 0.8184
[2021-03-20 14:10:48] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 14:10:48] Found 50000 unlabelled features.
[2021-03-20 14:11:00] Showing labelled: True (3338/50000 visible, 62 redundant)
[2021-03-20 14:11:00] Creating trainer with model on device: cuda
[2021-03-20 14:11:00] Training vgg-pretrained across 3400 data points in cifar10-augmented...
[2021-03-20 14:11:53] Training accuracy: 0.9921
[2021-03-20 14:11:53] Testing on 10000 data points...
[2021-03-20 14:11:55] Test score for 3400 training labels: 0.8250
[2021-03-20 14:11:55] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 14:11:55] Found 50000 unlabelled features.
[2021-03-20 14:12:08] Showing labelled: True (3730/50000 visible, 70 redundant)
[2021-03-20 14:12:08] Creating trainer with model on device: cuda
[2021-03-20 14:12:08] Training vgg-pretrained across 3800 data points in cifar10-augmented...
[2021-03-20 14:13:08] Training accuracy: 0.9914
[2021-03-20 14:13:08] Testing on 10000 data points...
[2021-03-20 14:13:10] Test score for 3800 training labels: 0.8303
[2021-03-20 14:13:10] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 14:13:10] Found 50000 unlabelled features.
[2021-03-20 14:13:22] Showing labelled: True (4124/50000 visible, 76 redundant)
[2021-03-20 14:13:22] Creating trainer with model on device: cuda
[2021-03-20 14:13:22] Training vgg-pretrained across 4200 data points in cifar10-augmented...
[2021-03-20 14:14:28] Training accuracy: 0.9945
[2021-03-20 14:14:28] Testing on 10000 data points...
[2021-03-20 14:14:30] Test score for 4200 training labels: 0.8406
[2021-03-20 14:14:30] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 14:14:30] Found 50000 unlabelled features.
[2021-03-20 14:14:42] Showing labelled: True (4518/50000 visible, 82 redundant)
[2021-03-20 14:14:42] Creating trainer with model on device: cuda
[2021-03-20 14:14:42] Training vgg-pretrained across 4600 data points in cifar10-augmented...
[2021-03-20 14:15:56] Training accuracy: 0.9941
[2021-03-20 14:15:56] Testing on 10000 data points...
[2021-03-20 14:15:57] Test score for 4600 training labels: 0.8452
[2021-03-20 14:15:57] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 14:15:57] Found 50000 unlabelled features.
[2021-03-20 14:16:10] Showing labelled: True (4907/50000 visible, 93 redundant)
[2021-03-20 14:16:10] Creating trainer with model on device: cuda
[2021-03-20 14:16:10] Training vgg-pretrained across 5000 data points in cifar10-augmented...
[2021-03-20 14:17:30] Training accuracy: 0.9938
[2021-03-20 14:17:30] Testing on 10000 data points...
[2021-03-20 14:17:32] Test score for 5000 training labels: 0.8501
[2021-03-20 14:17:32] Experiment repeat 4/4
[2021-03-20 14:17:32] Seeded: 8
[2021-03-20 14:17:32] Using 2.00% labels of the dataset (1000/50000)
[2021-03-20 14:17:32] Randomly labelled 1000/50000
[2021-03-20 14:17:32] Showing labelled: True (1000/50000 visible, 0 redundant)
[2021-03-20 14:17:32] Seeded: 9
[2021-03-20 14:17:32] Running: experiment 0
[2021-03-20 14:17:32] Showing labelled: True (1000/50000 visible, 0 redundant)
[2021-03-20 14:17:32] Creating pretrained=True VGG16...
[2021-03-20 14:17:34] No parameter reset since we are using a pretrained model.
[2021-03-20 14:17:34] vgg-pretrained: initialized 15245130 parameters.
[2021-03-20 14:17:34] Creating trainer with model on device: cuda
[2021-03-20 14:17:34] Training vgg-pretrained across 1000 data points in cifar10-augmented...
[2021-03-20 14:18:38] Training accuracy: 0.9924
[2021-03-20 14:18:38] Testing on 10000 data points...
[2021-03-20 14:18:40] Test score for 1000 training labels: 0.7401
[2021-03-20 14:18:40] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 14:18:40] Found 50000 unlabelled features.
[2021-03-20 14:18:53] Showing labelled: True (1400/50000 visible, 0 redundant)
[2021-03-20 14:18:53] Creating trainer with model on device: cuda
[2021-03-20 14:18:53] Training vgg-pretrained across 1400 data points in cifar10-augmented...
[2021-03-20 14:19:15] Training accuracy: 0.9546
[2021-03-20 14:19:15] Testing on 10000 data points...
[2021-03-20 14:19:17] Test score for 1400 training labels: 0.7659
[2021-03-20 14:19:17] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 14:19:17] Found 50000 unlabelled features.
[2021-03-20 14:19:30] Showing labelled: True (1791/50000 visible, 9 redundant)
[2021-03-20 14:19:30] Creating trainer with model on device: cuda
[2021-03-20 14:19:30] Training vgg-pretrained across 1800 data points in cifar10-augmented...
[2021-03-20 14:19:58] Training accuracy: 0.9375
[2021-03-20 14:19:58] Testing on 10000 data points...
[2021-03-20 14:20:00] Test score for 1800 training labels: 0.7689
[2021-03-20 14:20:00] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 14:20:00] Found 50000 unlabelled features.
[2021-03-20 14:20:13] Showing labelled: True (2172/50000 visible, 28 redundant)
[2021-03-20 14:20:13] Creating trainer with model on device: cuda
[2021-03-20 14:20:13] Training vgg-pretrained across 2200 data points in cifar10-augmented...
[2021-03-20 14:20:49] Training accuracy: 0.9759
[2021-03-20 14:20:49] Testing on 10000 data points...
[2021-03-20 14:20:51] Test score for 2200 training labels: 0.7964
[2021-03-20 14:20:51] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 14:20:51] Found 50000 unlabelled features.
[2021-03-20 14:21:04] Showing labelled: True (2561/50000 visible, 39 redundant)
[2021-03-20 14:21:04] Creating trainer with model on device: cuda
[2021-03-20 14:21:04] Training vgg-pretrained across 2600 data points in cifar10-augmented...
[2021-03-20 14:21:53] Training accuracy: 0.9862
[2021-03-20 14:21:53] Testing on 10000 data points...
[2021-03-20 14:21:55] Test score for 2600 training labels: 0.8064
[2021-03-20 14:21:55] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 14:21:55] Found 50000 unlabelled features.
[2021-03-20 14:22:07] Showing labelled: True (2947/50000 visible, 53 redundant)
[2021-03-20 14:22:07] Creating trainer with model on device: cuda
[2021-03-20 14:22:07] Training vgg-pretrained across 3000 data points in cifar10-augmented...
[2021-03-20 14:22:55] Training accuracy: 0.9868
[2021-03-20 14:22:55] Testing on 10000 data points...
[2021-03-20 14:22:57] Test score for 3000 training labels: 0.8190
[2021-03-20 14:22:57] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 14:22:57] Found 50000 unlabelled features.
[2021-03-20 14:23:09] Showing labelled: True (3339/50000 visible, 61 redundant)
[2021-03-20 14:23:09] Creating trainer with model on device: cuda
[2021-03-20 14:23:09] Training vgg-pretrained across 3400 data points in cifar10-augmented...
[2021-03-20 14:24:03] Training accuracy: 0.9787
[2021-03-20 14:24:03] Testing on 10000 data points...
[2021-03-20 14:24:05] Test score for 3400 training labels: 0.8207
[2021-03-20 14:24:05] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 14:24:05] Found 50000 unlabelled features.
[2021-03-20 14:24:18] Showing labelled: True (3720/50000 visible, 80 redundant)
[2021-03-20 14:24:18] Creating trainer with model on device: cuda
[2021-03-20 14:24:18] Training vgg-pretrained across 3800 data points in cifar10-augmented...
[2021-03-20 14:25:19] Training accuracy: 0.9913
[2021-03-20 14:25:19] Testing on 10000 data points...
[2021-03-20 14:25:21] Test score for 3800 training labels: 0.8316
[2021-03-20 14:25:21] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 14:25:21] Found 50000 unlabelled features.
[2021-03-20 14:25:35] Showing labelled: True (4108/50000 visible, 92 redundant)
[2021-03-20 14:25:35] Creating trainer with model on device: cuda
[2021-03-20 14:25:35] Training vgg-pretrained across 4200 data points in cifar10-augmented...
[2021-03-20 14:26:40] Training accuracy: 0.9901
[2021-03-20 14:26:40] Testing on 10000 data points...
[2021-03-20 14:26:43] Test score for 4200 training labels: 0.8340
[2021-03-20 14:26:43] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 14:26:43] Found 50000 unlabelled features.
[2021-03-20 14:26:56] Showing labelled: True (4497/50000 visible, 103 redundant)
[2021-03-20 14:26:56] Creating trainer with model on device: cuda
[2021-03-20 14:26:56] Training vgg-pretrained across 4600 data points in cifar10-augmented...
[2021-03-20 14:28:08] Training accuracy: 0.9935
[2021-03-20 14:28:08] Testing on 10000 data points...
[2021-03-20 14:28:10] Test score for 4600 training labels: 0.8410
[2021-03-20 14:28:10] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 14:28:10] Found 50000 unlabelled features.
[2021-03-20 14:28:23] Showing labelled: True (4888/50000 visible, 112 redundant)
[2021-03-20 14:28:23] Creating trainer with model on device: cuda
[2021-03-20 14:28:23] Training vgg-pretrained across 5000 data points in cifar10-augmented...
[2021-03-20 14:29:43] Training accuracy: 0.9961
[2021-03-20 14:29:43] Testing on 10000 data points...
[2021-03-20 14:29:45] Test score for 5000 training labels: 0.8425
[2021-03-20 14:29:45] Updated results: ../results/cifar10-aug/vgg_pretrained_lc_nolsmooth/results.json
