[2021-03-20 22:09:17] Created experiment 0:
[2021-03-20 22:09:17]  - Model: vgg-pretrained
[2021-03-20 22:09:17]  - Acquisition function: lc-beam-pweighted-relconf-coreset
[2021-03-20 22:09:17] Created experiment 1:
[2021-03-20 22:09:17]  - Model: vgg-pretrained
[2021-03-20 22:09:17]  - Acquisition function: random
[2021-03-20 22:09:17] Created experiment 2:
[2021-03-20 22:09:17]  - Model: vgg-pretrained
[2021-03-20 22:09:17]  - Acquisition function: greedy-coreset
[2021-03-20 22:09:17] Created experiment 3:
[2021-03-20 22:09:17]  - Model: vgg-pretrained
[2021-03-20 22:09:17]  - Acquisition function: lc-beam-pweighted-coreset
[2021-03-20 22:09:17] Created experiment 4:
[2021-03-20 22:09:17]  - Model: vgg-pretrained
[2021-03-20 22:09:17]  - Acquisition function: lc-beam-pweighted-relconf-coreset
[2021-03-20 22:09:17] Loading cifar10 test set...
[2021-03-20 22:09:18] Experiment repeat 1/1
[2021-03-20 22:09:18] Seeded: 5
[2021-03-20 22:09:18] Using 0.10% labels of the dataset (50/50000)
[2021-03-20 22:09:18] Randomly labelled 50/50000
[2021-03-20 22:09:18] Showing labelled: True (50/50000 visible, 0 redundant)
[2021-03-20 22:09:18] Seeded: 6
[2021-03-20 22:09:18] Running: experiment 0
[2021-03-20 22:09:18] Showing labelled: True (50/50000 visible, 0 redundant)
[2021-03-20 22:09:18] Creating pretrained=True VGG16...
[2021-03-20 22:09:20] No parameter reset since we are using a pretrained model.
[2021-03-20 22:09:20] vgg-pretrained: initialized 15245130 parameters.
[2021-03-20 22:09:20] Creating trainer with model on device: cuda
[2021-03-20 22:09:25] Training vgg-pretrained across 50 data points in cifar10...
[2021-03-20 22:09:25] Training accuracy: 0.1000
[2021-03-20 22:09:25] Testing on 10000 data points...
[2021-03-20 22:09:27] Test score for 50 training labels: 0.0906
[2021-03-20 22:09:27] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 22:09:27] Found 50000 unlabelled features.
[2021-03-20 22:09:38] Computing distance between 50 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 22:09:38] Searching for coresets with 20 beams...
[2021-03-20 22:09:42] Showing labelled: True (100/50000 visible, 0 redundant)
[2021-03-20 22:09:42] Creating trainer with model on device: cuda
[2021-03-20 22:09:42] Training vgg-pretrained across 100 data points in cifar10...
[2021-03-20 22:09:42] Training accuracy: 0.0812
[2021-03-20 22:09:42] Testing on 10000 data points...
[2021-03-20 22:09:44] Test score for 100 training labels: 0.0905
[2021-03-20 22:09:44] Seeded: 6
[2021-03-20 22:09:44] Running: experiment 1
[2021-03-20 22:09:44] Showing labelled: True (50/50000 visible, 0 redundant)
[2021-03-20 22:09:44] Creating pretrained=True VGG16...
[2021-03-20 22:09:45] No parameter reset since we are using a pretrained model.
[2021-03-20 22:09:45] vgg-pretrained: initialized 15245130 parameters.
[2021-03-20 22:09:45] Creating trainer with model on device: cuda
[2021-03-20 22:09:45] Training vgg-pretrained across 50 data points in cifar10...
[2021-03-20 22:09:45] Training accuracy: 0.1000
[2021-03-20 22:09:45] Testing on 10000 data points...
[2021-03-20 22:09:47] Test score for 50 training labels: 0.0906
[2021-03-20 22:09:47] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 22:09:47] Found 50000 unlabelled features.
[2021-03-20 22:09:53] Showing labelled: True (100/50000 visible, 0 redundant)
[2021-03-20 22:09:53] Creating trainer with model on device: cuda
[2021-03-20 22:09:53] Training vgg-pretrained across 100 data points in cifar10...
[2021-03-20 22:09:53] Training accuracy: 0.0608
[2021-03-20 22:09:53] Testing on 10000 data points...
[2021-03-20 22:09:55] Test score for 100 training labels: 0.0905
[2021-03-20 22:09:55] Seeded: 6
[2021-03-20 22:09:55] Running: experiment 2
[2021-03-20 22:09:55] Showing labelled: True (50/50000 visible, 0 redundant)
[2021-03-20 22:09:55] Creating pretrained=True VGG16...
[2021-03-20 22:09:56] No parameter reset since we are using a pretrained model.
[2021-03-20 22:09:56] vgg-pretrained: initialized 15245130 parameters.
[2021-03-20 22:09:56] Creating trainer with model on device: cuda
[2021-03-20 22:09:56] Training vgg-pretrained across 50 data points in cifar10...
[2021-03-20 22:09:56] Training accuracy: 0.1000
[2021-03-20 22:09:56] Testing on 10000 data points...
[2021-03-20 22:09:58] Test score for 50 training labels: 0.0906
[2021-03-20 22:09:58] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 22:09:58] Found 50000 unlabelled features.
[2021-03-20 22:10:09] Computing distance between 50 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 22:10:10] Searching for coresets greedily...
[2021-03-20 22:10:10] Showing labelled: True (100/50000 visible, 0 redundant)
[2021-03-20 22:10:10] Creating trainer with model on device: cuda
[2021-03-20 22:10:10] Training vgg-pretrained across 100 data points in cifar10...
[2021-03-20 22:10:10] Training accuracy: 0.0635
[2021-03-20 22:10:10] Testing on 10000 data points...
[2021-03-20 22:10:12] Test score for 100 training labels: 0.0907
[2021-03-20 22:10:12] Seeded: 6
[2021-03-20 22:10:12] Running: experiment 3
[2021-03-20 22:10:12] Showing labelled: True (50/50000 visible, 0 redundant)
[2021-03-20 22:10:12] Creating pretrained=True VGG16...
[2021-03-20 22:10:14] No parameter reset since we are using a pretrained model.
[2021-03-20 22:10:14] vgg-pretrained: initialized 15245130 parameters.
[2021-03-20 22:10:14] Creating trainer with model on device: cuda
[2021-03-20 22:10:14] Training vgg-pretrained across 50 data points in cifar10...
[2021-03-20 22:10:14] Training accuracy: 0.1000
[2021-03-20 22:10:14] Testing on 10000 data points...
[2021-03-20 22:10:16] Test score for 50 training labels: 0.0906
[2021-03-20 22:10:16] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 22:10:16] Found 50000 unlabelled features.
[2021-03-20 22:10:26] Computing distance between 50 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 22:10:27] Searching for coresets with 20 beams...
[2021-03-20 22:10:30] Showing labelled: True (100/50000 visible, 0 redundant)
[2021-03-20 22:10:30] Creating trainer with model on device: cuda
[2021-03-20 22:10:30] Training vgg-pretrained across 100 data points in cifar10...
[2021-03-20 22:10:30] Training accuracy: 0.0784
[2021-03-20 22:10:30] Testing on 10000 data points...
[2021-03-20 22:10:32] Test score for 100 training labels: 0.0907
[2021-03-20 22:10:32] Seeded: 6
[2021-03-20 22:10:32] Running: experiment 4
[2021-03-20 22:10:32] Showing labelled: True (50/50000 visible, 0 redundant)
[2021-03-20 22:10:32] Creating pretrained=True VGG16...
[2021-03-20 22:10:34] No parameter reset since we are using a pretrained model.
[2021-03-20 22:10:34] vgg-pretrained: initialized 15245130 parameters.
[2021-03-20 22:10:34] Creating trainer with model on device: cuda
[2021-03-20 22:10:34] Training vgg-pretrained across 50 data points in cifar10...
[2021-03-20 22:10:34] Training accuracy: 0.1000
[2021-03-20 22:10:34] Testing on 10000 data points...
[2021-03-20 22:10:36] Test score for 50 training labels: 0.0906
[2021-03-20 22:10:36] Showing labelled: False (50000/50000 visible, 0 redundant)
[2021-03-20 22:10:36] Found 50000 unlabelled features.
[2021-03-20 22:10:46] Computing distance between 50 labelled and 50000 unlabelled vectors of length 512...
[2021-03-20 22:10:47] Searching for coresets with 1 beams...
[2021-03-20 22:10:47] Showing labelled: True (100/50000 visible, 0 redundant)
[2021-03-20 22:10:47] Creating trainer with model on device: cuda
[2021-03-20 22:10:47] Training vgg-pretrained across 100 data points in cifar10...
[2021-03-20 22:10:47] Training accuracy: 0.0960
[2021-03-20 22:10:47] Testing on 10000 data points...
[2021-03-20 22:10:49] Test score for 100 training labels: 0.0903
[2021-03-20 22:10:49] Updated results: ../results/cifar10/5k-steps-test/results.json
