[2021-03-20 17:35:40] Created experiment 0:
[2021-03-20 17:35:40]  - Model: vgg-pretrained
[2021-03-20 17:35:40]  - Acquisition function: lc-beam-pweighted-coreset
[2021-03-20 17:35:40] Loading cifar10 test set...
[2021-03-20 17:35:41] Experiment repeat 1/1
[2021-03-20 17:35:41] Seeded: 5
[2021-03-20 17:35:41] Using 2.00% labels of the dataset (1000/50000)
[2021-03-20 17:35:41] Randomly labelled 1000/50000
[2021-03-20 17:35:41] Showing labelled: True (1000/50000 visible, 0 redundant)
[2021-03-20 17:35:41] Seeded: 6
[2021-03-20 17:35:41] Running: experiment 0
[2021-03-20 17:35:41] Showing labelled: True (1000/50000 visible, 0 redundant)
[2021-03-20 17:35:41] Creating pretrained=True VGG16...
[2021-03-20 17:35:43] No parameter reset since we are using a pretrained model.
[2021-03-20 17:35:43] vgg-pretrained: initialized 15245130 parameters.
[2021-03-20 17:35:43] Creating trainer with model on device: cuda
[2021-03-20 17:35:47] Training vgg-pretrained across 1000 data points in cifar10...
[2021-03-20 17:36:09] Training accuracy: 0.9858
[2021-03-20 17:36:09] Testing on 10000 data points...
[2021-03-20 17:36:11] Test score for 1000 training labels: 0.7172
[2021-03-20 17:36:11] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 17:36:11] Found 50000 unlabelled features.
[2021-03-20 17:36:21] Computing distance between 1000 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 17:36:24] Searching for coresets with 20 beams...
[2021-03-20 17:36:52] Showing labelled: True (1400/50000 visible, 0 redundant)
[2021-03-20 17:36:52] Creating trainer with model on device: cuda
[2021-03-20 17:36:52] Training vgg-pretrained across 1400 data points in cifar10...
[2021-03-20 17:37:00] Training accuracy: 0.9496
[2021-03-20 17:37:00] Testing on 10000 data points...
[2021-03-20 17:37:02] Test score for 1400 training labels: 0.7461
[2021-03-20 17:37:02] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 17:37:02] Found 50000 unlabelled features.
[2021-03-20 17:37:12] Computing distance between 1400 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 17:37:16] Searching for coresets with 20 beams...
[2021-03-20 17:37:44] Showing labelled: True (1800/50000 visible, 0 redundant)
[2021-03-20 17:37:44] Creating trainer with model on device: cuda
[2021-03-20 17:37:44] Training vgg-pretrained across 1800 data points in cifar10...
[2021-03-20 17:37:54] Training accuracy: 0.9492
[2021-03-20 17:37:54] Testing on 10000 data points...
[2021-03-20 17:37:56] Test score for 1800 training labels: 0.7630
[2021-03-20 17:37:56] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 17:37:56] Found 50000 unlabelled features.
[2021-03-20 17:38:06] Computing distance between 1800 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 17:38:11] Searching for coresets with 20 beams...
[2021-03-20 17:38:39] Showing labelled: True (2200/50000 visible, 0 redundant)
[2021-03-20 17:38:39] Creating trainer with model on device: cuda
[2021-03-20 17:38:39] Training vgg-pretrained across 2200 data points in cifar10...
[2021-03-20 17:38:53] Training accuracy: 0.9865
[2021-03-20 17:38:53] Testing on 10000 data points...
[2021-03-20 17:38:55] Test score for 2200 training labels: 0.7832
[2021-03-20 17:38:55] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 17:38:55] Found 50000 unlabelled features.
[2021-03-20 17:39:05] Computing distance between 2200 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 17:39:11] Searching for coresets with 20 beams...
[2021-03-20 17:39:39] Showing labelled: True (2600/50000 visible, 0 redundant)
[2021-03-20 17:39:39] Creating trainer with model on device: cuda
[2021-03-20 17:39:39] Training vgg-pretrained across 2600 data points in cifar10...
[2021-03-20 17:39:55] Training accuracy: 0.9675
[2021-03-20 17:39:55] Testing on 10000 data points...
[2021-03-20 17:39:56] Test score for 2600 training labels: 0.7897
[2021-03-20 17:39:56] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 17:39:56] Found 50000 unlabelled features.
[2021-03-20 17:40:07] Computing distance between 2600 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 17:40:14] Searching for coresets with 20 beams...
[2021-03-20 17:40:42] Showing labelled: True (3000/50000 visible, 0 redundant)
[2021-03-20 17:40:42] Creating trainer with model on device: cuda
[2021-03-20 17:40:42] Training vgg-pretrained across 3000 data points in cifar10...
[2021-03-20 17:41:00] Training accuracy: 0.9979
[2021-03-20 17:41:00] Testing on 10000 data points...
[2021-03-20 17:41:02] Test score for 3000 training labels: 0.8022
[2021-03-20 17:41:02] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 17:41:02] Found 50000 unlabelled features.
[2021-03-20 17:41:12] Computing distance between 3000 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 17:41:20] Searching for coresets with 20 beams...
[2021-03-20 17:41:48] Showing labelled: True (3400/50000 visible, 0 redundant)
[2021-03-20 17:41:48] Creating trainer with model on device: cuda
[2021-03-20 17:41:48] Training vgg-pretrained across 3400 data points in cifar10...
[2021-03-20 17:42:09] Training accuracy: 0.9978
[2021-03-20 17:42:09] Testing on 10000 data points...
[2021-03-20 17:42:11] Test score for 3400 training labels: 0.8148
[2021-03-20 17:42:11] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 17:42:11] Found 50000 unlabelled features.
[2021-03-20 17:42:21] Computing distance between 3400 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 17:42:30] Searching for coresets with 20 beams...
[2021-03-20 17:42:59] Showing labelled: True (3800/50000 visible, 0 redundant)
[2021-03-20 17:42:59] Creating trainer with model on device: cuda
[2021-03-20 17:42:59] Training vgg-pretrained across 3800 data points in cifar10...
[2021-03-20 17:43:22] Training accuracy: 0.9988
[2021-03-20 17:43:22] Testing on 10000 data points...
[2021-03-20 17:43:24] Test score for 3800 training labels: 0.8213
[2021-03-20 17:43:24] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 17:43:24] Found 50000 unlabelled features.
[2021-03-20 17:43:34] Computing distance between 3800 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 17:43:45] Searching for coresets with 20 beams...
[2021-03-20 17:44:13] Showing labelled: True (4200/50000 visible, 0 redundant)
[2021-03-20 17:44:13] Creating trainer with model on device: cuda
[2021-03-20 17:44:13] Training vgg-pretrained across 4200 data points in cifar10...
[2021-03-20 17:44:38] Training accuracy: 0.9976
[2021-03-20 17:44:38] Testing on 10000 data points...
[2021-03-20 17:44:40] Test score for 4200 training labels: 0.8275
[2021-03-20 17:44:40] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 17:44:40] Found 50000 unlabelled features.
[2021-03-20 17:44:51] Computing distance between 4200 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 17:45:02] Searching for coresets with 20 beams...
[2021-03-20 17:45:30] Showing labelled: True (4600/50000 visible, 0 redundant)
[2021-03-20 17:45:30] Creating trainer with model on device: cuda
[2021-03-20 17:45:30] Training vgg-pretrained across 4600 data points in cifar10...
[2021-03-20 17:45:58] Training accuracy: 0.9971
[2021-03-20 17:45:58] Testing on 10000 data points...
[2021-03-20 17:46:00] Test score for 4600 training labels: 0.8348
[2021-03-20 17:46:00] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 17:46:00] Found 50000 unlabelled features.
[2021-03-20 17:46:10] Computing distance between 4600 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 17:46:23] Searching for coresets with 20 beams...
[2021-03-20 17:46:51] Showing labelled: True (5000/50000 visible, 0 redundant)
[2021-03-20 17:46:51] Creating trainer with model on device: cuda
[2021-03-20 17:46:51] Training vgg-pretrained across 5000 data points in cifar10...
[2021-03-20 17:47:21] Training accuracy: 0.9994
[2021-03-20 17:47:21] Testing on 10000 data points...
[2021-03-20 17:47:23] Test score for 5000 training labels: 0.8322
[2021-03-20 17:47:23] Updated results: ../results/cifar10/vgg_pretrained_pweighted/results.json
