Commits · 45a6f2edd920765aad5ef46e5c06330d95170c8c · chenpangpang / transformers

24 Sep, 2019 15 commits
- docstring for GPT · 45a6f2ed
  thomwolf authored Sep 24, 2019
  
  45a6f2ed
- docstring for GPT2 · e7ba5bc8
  thomwolf authored Sep 24, 2019
  
  e7ba5bc8
- distilbert docstring · b94f73ba
  thomwolf authored Sep 24, 2019
  
  b94f73ba
- docstrings for bert · 9678c494
  thomwolf authored Sep 24, 2019
  
  9678c494
- fix imports · f3d1511b
  thomwolf authored Sep 24, 2019
  
  f3d1511b
- update automodels · dd2d90f3
  thomwolf authored Sep 24, 2019
  
  dd2d90f3
- add save_pretrained · ee261439
  thomwolf authored Sep 24, 2019
  
  ee261439
- double loading ok · 29bb3e4e
  thomwolf authored Sep 24, 2019
  
  29bb3e4e
- update loading logics · f5397ffc
  thomwolf authored Sep 24, 2019
  
  f5397ffc
- updating to load tf model in pt - fixing headmasking test · 271f2136
  thomwolf authored Sep 24, 2019
  
  271f2136
- fix tests chen only using tf · cf9c1cbb
  thomwolf authored Sep 24, 2019
  
  cf9c1cbb
- update circleCi · 2167e366
  thomwolf authored Sep 24, 2019
  
  2167e366
- bidirectional conversion TF <=> PT - extended tests · e9a103c1
  thomwolf authored Sep 24, 2019
  
  e9a103c1
- converting distilled/fine-tuned models · a7e01a24
  thomwolf authored Sep 24, 2019
  
  a7e01a24
- fix roberta conversion script · 8ba44ced
  thomwolf authored Sep 24, 2019
  
  8ba44ced
23 Sep, 2019 10 commits
- update __init__ and conversion script · 2b11fa51
  thomwolf authored Sep 23, 2019
  
  2b11fa51
- fix roberta test · 6448396d
  thomwolf authored Sep 23, 2019
  
  6448396d
- Merge branch 'tf2' of https://github.com/huggingface/pytorch-transformers into tf2 · 1e47dee2
  thomwolf authored Sep 23, 2019
  
  1e47dee2
- updated models input format + tests · c9591f6f
  thomwolf authored Sep 23, 2019
  
  c9591f6f
- Fix TFBert tests in Python 3.5 · 798da627
  Julien Chaumond authored Sep 23, 2019
  
  798da627
- fix the skipping · c014d1f0
  thomwolf authored Sep 23, 2019
  
  c014d1f0
- skipping pretrained TF model tests for now · 0b22e47a
  thomwolf authored Sep 23, 2019
  
  0b22e47a
- test circleCI h5py version · 830d212b
  thomwolf authored Sep 23, 2019
  
  830d212b
- fix XLM tests · a31e591d
  thomwolf authored Sep 23, 2019
  
  a31e591d
- tests for distilbert and roberta · 447de34d
  thomwolf authored Sep 23, 2019
  
  447de34d
20 Sep, 2019 1 commit
- roberta and distilbert · 68a3e022
  thomwolf authored Sep 20, 2019
  
  68a3e022
18 Sep, 2019 6 commits
- OpenAI GPT tests ok · 3a527fa8
  thomwolf authored Sep 18, 2019
  
  3a527fa8
- hot fix · 556442af
  thomwolf authored Sep 18, 2019
  
  556442af
- fix xlm lang_embeddings loading · 160b5d60
  thomwolf authored Sep 18, 2019
  
  160b5d60
- fix tests · 26497d11
  thomwolf authored Sep 18, 2019
  
  26497d11
- update pt-tf conversion script · 6a083fd4
  thomwolf authored Sep 18, 2019
  
  6a083fd4
- upgrade max model difference to 2e-2 (for transfo-xl adaptive softmax + inputs) · f6969cc1
  thomwolf authored Sep 18, 2019
  
  f6969cc1
13 Sep, 2019 3 commits
- add layer_norm_epsilon configuration for transformer xl · 4b956b2a
  thomwolf authored Sep 13, 2019
  
  4b956b2a
- skip finetuned checkpoints · b97af8cc
  thomwolf authored Sep 13, 2019
  
  b97af8cc
- adding TF 2.0 adaptive softmax with logits + loss outputs · 65c49bb2
  thomwolf authored Sep 13, 2019
  
  65c49bb2
12 Sep, 2019 4 commits
- fix · 39c38b2e
  thomwolf authored Sep 12, 2019
  
  39c38b2e
- fix bert layernorm · dcddf498
  thomwolf authored Sep 12, 2019
  
  dcddf498
- clean up cache after conversion · d3a3a035
  thomwolf authored Sep 12, 2019
  
  d3a3a035
- convert all models · a84adddd
  thomwolf authored Sep 12, 2019
  
  a84adddd
11 Sep, 2019 1 commit
- XLMWithLMHead fixed - standardize conversion · 969d3ae9
  thomwolf authored Sep 11, 2019
  
  969d3ae9