Commit
•
14a227a
1
Parent(s):
0ce0079
added training log
Browse files- training.log +56 -0
training.log
ADDED
|
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2021-05-21 19:52:16,903 INFO __main__ Namespace(adjust_lr=False, config='torchdistill/configs/sample/glue/cola/ce/bert_large_uncased.yaml', log='log/glue/cola/ce/bert_large_uncased.txt', private_output='leaderboard/glue/standard/bert_large_uncased/', seed=None, student_only=False, task_name='cola', test_only=False, world_size=1)
|
| 2 |
+
2021-05-21 19:52:16,964 INFO __main__ Distributed environment: NO
|
| 3 |
+
Num processes: 1
|
| 4 |
+
Process index: 0
|
| 5 |
+
Local process index: 0
|
| 6 |
+
Device: cuda
|
| 7 |
+
Use FP16 precision: True
|
| 8 |
+
|
| 9 |
+
2021-05-21 19:52:17,253 INFO filelock Lock 139707469705104 acquired on /root/.cache/huggingface/transformers/1cf090f220f9674b67b3434decfe4d40a6532d7849653eac435ff94d31a4904c.1d03e5e4fa2db2532c517b2cd98290d8444b237619bd3d2039850a6d5e86473d.lock
|
| 10 |
+
2021-05-21 19:52:17,525 INFO filelock Lock 139707469705104 released on /root/.cache/huggingface/transformers/1cf090f220f9674b67b3434decfe4d40a6532d7849653eac435ff94d31a4904c.1d03e5e4fa2db2532c517b2cd98290d8444b237619bd3d2039850a6d5e86473d.lock
|
| 11 |
+
2021-05-21 19:52:18,060 INFO filelock Lock 139707469840208 acquired on /root/.cache/huggingface/transformers/e12f02d630da91a0982ce6db1ad595231d155a2b725ab106971898276d842ecc.d789d64ebfe299b0e416afc4a169632f903f693095b4629a7ea271d5a0cf2c99.lock
|
| 12 |
+
2021-05-21 19:52:18,665 INFO filelock Lock 139707469840208 released on /root/.cache/huggingface/transformers/e12f02d630da91a0982ce6db1ad595231d155a2b725ab106971898276d842ecc.d789d64ebfe299b0e416afc4a169632f903f693095b4629a7ea271d5a0cf2c99.lock
|
| 13 |
+
2021-05-21 19:52:18,932 INFO filelock Lock 139707469840208 acquired on /root/.cache/huggingface/transformers/475d46024228961ca8770cead39e1079f135fd2441d14cf216727ffac8d41d78.7f2721073f19841be16f41b0a70b600ca6b880c8f3df6f3535cbc704371bdfa4.lock
|
| 14 |
+
2021-05-21 19:52:19,623 INFO filelock Lock 139707469840208 released on /root/.cache/huggingface/transformers/475d46024228961ca8770cead39e1079f135fd2441d14cf216727ffac8d41d78.7f2721073f19841be16f41b0a70b600ca6b880c8f3df6f3535cbc704371bdfa4.lock
|
| 15 |
+
2021-05-21 19:52:20,432 INFO filelock Lock 139707469872272 acquired on /root/.cache/huggingface/transformers/300ecd79785b4602752c0085f8a89c3f0232ef367eda291c79a5600f3778b677.20430bd8e10ef77a7d2977accefe796051e01bc2fc4aa146bc862997a1a15e79.lock
|
| 16 |
+
2021-05-21 19:52:20,702 INFO filelock Lock 139707469872272 released on /root/.cache/huggingface/transformers/300ecd79785b4602752c0085f8a89c3f0232ef367eda291c79a5600f3778b677.20430bd8e10ef77a7d2977accefe796051e01bc2fc4aa146bc862997a1a15e79.lock
|
| 17 |
+
2021-05-21 19:52:20,993 INFO filelock Lock 139707502187088 acquired on /root/.cache/huggingface/transformers/1d959166dd7e047e57ea1b2d9b7b9669938a7e90c5e37a03961ad9f15eaea17f.fea64cd906e3766b04c92397f9ad3ff45271749cbe49829a079dd84e34c1697d.lock
|
| 18 |
+
2021-05-21 19:52:43,690 INFO filelock Lock 139707502187088 released on /root/.cache/huggingface/transformers/1d959166dd7e047e57ea1b2d9b7b9669938a7e90c5e37a03961ad9f15eaea17f.fea64cd906e3766b04c92397f9ad3ff45271749cbe49829a079dd84e34c1697d.lock
|
| 19 |
+
2021-05-21 19:52:53,493 INFO __main__ Start training
|
| 20 |
+
2021-05-21 19:52:53,493 INFO torchdistill.models.util [student model]
|
| 21 |
+
2021-05-21 19:52:53,493 INFO torchdistill.models.util Using the original student model
|
| 22 |
+
2021-05-21 19:52:53,493 INFO torchdistill.core.training Loss = 1.0 * OrgLoss
|
| 23 |
+
2021-05-21 19:52:59,798 INFO torchdistill.misc.log Epoch: [0] [ 0/268] eta: 0:02:14 lr: 1.9975124378109453e-05 sample/s: 8.285208596713021 loss: 0.7285 (0.7285) time: 0.5037 data: 0.0209 max mem: 5355
|
| 24 |
+
2021-05-21 19:53:18,618 INFO torchdistill.misc.log Epoch: [0] [ 50/268] eta: 0:01:22 lr: 1.873134328358209e-05 sample/s: 11.48804134737782 loss: 0.5990 (0.6344) time: 0.3712 data: 0.0031 max mem: 7403
|
| 25 |
+
2021-05-21 19:53:37,548 INFO torchdistill.misc.log Epoch: [0] [100/268] eta: 0:01:03 lr: 1.7487562189054726e-05 sample/s: 9.439338529019249 loss: 0.4860 (0.5769) time: 0.3813 data: 0.0029 max mem: 7403
|
| 26 |
+
2021-05-21 19:53:56,330 INFO torchdistill.misc.log Epoch: [0] [150/268] eta: 0:00:44 lr: 1.6243781094527366e-05 sample/s: 11.48365361250954 loss: 0.4258 (0.5312) time: 0.3704 data: 0.0029 max mem: 7403
|
| 27 |
+
2021-05-21 19:54:14,604 INFO torchdistill.misc.log Epoch: [0] [200/268] eta: 0:00:25 lr: 1.5000000000000002e-05 sample/s: 9.44590729538729 loss: 0.3916 (0.4993) time: 0.3633 data: 0.0030 max mem: 7403
|
| 28 |
+
2021-05-21 19:54:33,394 INFO torchdistill.misc.log Epoch: [0] [250/268] eta: 0:00:06 lr: 1.3756218905472638e-05 sample/s: 8.100720542983192 loss: 0.4126 (0.4791) time: 0.3930 data: 0.0030 max mem: 7403
|
| 29 |
+
2021-05-21 19:54:39,779 INFO torchdistill.misc.log Epoch: [0] Total time: 0:01:40
|
| 30 |
+
2021-05-21 19:54:42,894 INFO /usr/local/lib/python3.7/dist-packages/datasets/metric.py Removing /root/.cache/huggingface/metrics/glue/cola/default_experiment-1-0.arrow
|
| 31 |
+
2021-05-21 19:54:42,894 INFO __main__ Validation: matthews_correlation = 0.5755578350293411
|
| 32 |
+
2021-05-21 19:54:42,894 INFO __main__ Updating ckpt
|
| 33 |
+
2021-05-21 19:54:47,852 INFO torchdistill.misc.log Epoch: [1] [ 0/268] eta: 0:01:37 lr: 1.3308457711442788e-05 sample/s: 11.084181081356062 loss: 0.2307 (0.2307) time: 0.3650 data: 0.0042 max mem: 7826
|
| 34 |
+
2021-05-21 19:55:06,275 INFO torchdistill.misc.log Epoch: [1] [ 50/268] eta: 0:01:20 lr: 1.2064676616915423e-05 sample/s: 11.449492398950676 loss: 0.2716 (0.2880) time: 0.3855 data: 0.0030 max mem: 7826
|
| 35 |
+
2021-05-21 19:55:25,586 INFO torchdistill.misc.log Epoch: [1] [100/268] eta: 0:01:03 lr: 1.082089552238806e-05 sample/s: 11.479088581543106 loss: 0.2913 (0.2842) time: 0.3964 data: 0.0030 max mem: 7829
|
| 36 |
+
2021-05-21 19:55:44,446 INFO torchdistill.misc.log Epoch: [1] [150/268] eta: 0:00:44 lr: 9.577114427860697e-06 sample/s: 11.456959277340616 loss: 0.2614 (0.2788) time: 0.3965 data: 0.0031 max mem: 7829
|
| 37 |
+
2021-05-21 19:56:02,782 INFO torchdistill.misc.log Epoch: [1] [200/268] eta: 0:00:25 lr: 8.333333333333334e-06 sample/s: 11.499151471285733 loss: 0.2638 (0.2769) time: 0.3481 data: 0.0030 max mem: 7829
|
| 38 |
+
2021-05-21 19:56:21,192 INFO torchdistill.misc.log Epoch: [1] [250/268] eta: 0:00:06 lr: 7.089552238805971e-06 sample/s: 11.464639589773867 loss: 0.2114 (0.2705) time: 0.3626 data: 0.0029 max mem: 7829
|
| 39 |
+
2021-05-21 19:56:27,657 INFO torchdistill.misc.log Epoch: [1] Total time: 0:01:40
|
| 40 |
+
2021-05-21 19:56:30,765 INFO /usr/local/lib/python3.7/dist-packages/datasets/metric.py Removing /root/.cache/huggingface/metrics/glue/cola/default_experiment-1-0.arrow
|
| 41 |
+
2021-05-21 19:56:30,766 INFO __main__ Validation: matthews_correlation = 0.5534186082067715
|
| 42 |
+
2021-05-21 19:56:31,119 INFO torchdistill.misc.log Epoch: [2] [ 0/268] eta: 0:01:34 lr: 6.64179104477612e-06 sample/s: 11.464655258421223 loss: 0.1139 (0.1139) time: 0.3523 data: 0.0034 max mem: 7829
|
| 43 |
+
2021-05-21 19:56:49,610 INFO torchdistill.misc.log Epoch: [2] [ 50/268] eta: 0:01:20 lr: 5.398009950248757e-06 sample/s: 9.450781311612081 loss: 0.1094 (0.1528) time: 0.3744 data: 0.0030 max mem: 7829
|
| 44 |
+
2021-05-21 19:57:08,112 INFO torchdistill.misc.log Epoch: [2] [100/268] eta: 0:01:02 lr: 4.1542288557213935e-06 sample/s: 11.468636482331405 loss: 0.1260 (0.1527) time: 0.3559 data: 0.0030 max mem: 7829
|
| 45 |
+
2021-05-21 19:57:26,908 INFO torchdistill.misc.log Epoch: [2] [150/268] eta: 0:00:43 lr: 2.9104477611940303e-06 sample/s: 11.340868578891262 loss: 0.1274 (0.1475) time: 0.3707 data: 0.0031 max mem: 7829
|
| 46 |
+
2021-05-21 19:57:45,674 INFO torchdistill.misc.log Epoch: [2] [200/268] eta: 0:00:25 lr: 1.6666666666666667e-06 sample/s: 9.455734970044356 loss: 0.1542 (0.1513) time: 0.3704 data: 0.0030 max mem: 7829
|
| 47 |
+
2021-05-21 19:58:04,690 INFO torchdistill.misc.log Epoch: [2] [250/268] eta: 0:00:06 lr: 4.2288557213930354e-07 sample/s: 11.481366714661908 loss: 0.0847 (0.1528) time: 0.3822 data: 0.0031 max mem: 7829
|
| 48 |
+
2021-05-21 19:58:10,794 INFO torchdistill.misc.log Epoch: [2] Total time: 0:01:40
|
| 49 |
+
2021-05-21 19:58:13,901 INFO /usr/local/lib/python3.7/dist-packages/datasets/metric.py Removing /root/.cache/huggingface/metrics/glue/cola/default_experiment-1-0.arrow
|
| 50 |
+
2021-05-21 19:58:13,902 INFO __main__ Validation: matthews_correlation = 0.6008507462939041
|
| 51 |
+
2021-05-21 19:58:13,902 INFO __main__ Updating ckpt
|
| 52 |
+
2021-05-21 19:58:24,318 INFO __main__ [Student: bert-large-uncased]
|
| 53 |
+
2021-05-21 19:58:27,448 INFO /usr/local/lib/python3.7/dist-packages/datasets/metric.py Removing /root/.cache/huggingface/metrics/glue/cola/default_experiment-1-0.arrow
|
| 54 |
+
2021-05-21 19:58:27,449 INFO __main__ Test: matthews_correlation = 0.6008507462939041
|
| 55 |
+
2021-05-21 19:58:27,449 INFO __main__ Start prediction for private dataset(s)
|
| 56 |
+
2021-05-21 19:58:27,450 INFO __main__ cola/test: 1063 samples
|