Default Branch

ae333d04b2 · torch.cuda.is_available() is redundant as apex handles that internally (#9350) · Updated 2020-12-30 17:09:51 +08:00

Branches

dd0e7c3995 · Uncomment useful code. · Updated 2020-10-31 02:12:43 +08:00    c-my

499
2
Compare

95ac70c5af · Removing dead code. · Updated 2020-10-30 22:38:46 +08:00    c-my

503
3
Compare

37b5af807c · Fixing docs with their rewritten version. · Updated 2020-10-30 17:49:15 +08:00    c-my

503
2
Compare

869ec78cdb · Fixing doc style. · Updated 2020-10-27 21:53:47 +08:00    c-my

553
4
Compare

9ebbe55bc3 · Fixup · Updated 2020-10-27 09:35:37 +08:00    c-my

556
42
Compare

b9b777749b · Tentative improvement on sequence_length error for position_ids · Updated 2020-10-26 18:01:16 +08:00    c-my

579
1
Compare

16a10cb6c2 · adding fast-slow equivalence tests · Updated 2020-10-23 22:13:27 +08:00    c-my

594
1
Compare

ac02f7530d · Correct cache · Updated 2020-10-16 19:33:29 +08:00    c-my

708
8
Compare

04cc1982f3 · refactor encoder layer · Updated 2020-10-14 01:39:02 +08:00    c-my

756
36
Compare

d8100a7fb2 · Actually changing the default resolution order to account for model · Updated 2020-10-12 18:54:45 +08:00    c-my

735
2
Compare

1d80765b8f · save intermediate · Updated 2020-10-10 01:00:44 +08:00    c-my

756
32
Compare

96ecc23fd7 · Complete instruction · Updated 2020-10-10 00:09:06 +08:00    c-my

746
2
Compare

76e05518bb · Better model links in the README and index · Updated 2020-10-09 22:54:40 +08:00    c-my

747
0
Included

b173ea6b2e · save_copy · Updated 2020-10-07 02:47:34 +08:00    c-my

877
18
Compare

30e2ce2d5b · splitting fast and slow tokenizers [WIP] · Updated 2020-10-06 17:29:20 +08:00    c-my

791
42
Compare

51c36e4921 · Trigger slow tests · Updated 2020-10-05 21:31:51 +08:00    c-my

810
21
Compare

7cfdbb24b7 · PyTorch initial token classification pipeline. · Updated 2020-10-02 18:34:31 +08:00    c-my

928
3
Compare

34eb3fcd55 · Fix automodel test · Updated 2020-09-28 13:28:17 +08:00    c-my

882
168
Compare

7dfdf793bb · Fixing case in which Trainer hung while saving model in distributed training (#7365) · Updated 2020-09-24 21:56:40 +08:00    c-my

909
0
Included

c1e44a3147 · boom boom · Updated 2020-09-22 08:12:16 +08:00    c-my

955
10
Compare