patil-suraj
Follow
Pinned
473 contributions in the last year
Less
More
Activity overview
Contributed to
huggingface/transformers,
huggingface/datasets,
patil-suraj/onnx_transformers
and 5 other
repositories
Contribution activity
February 2021
Created 13 commits in 2 repositories
Created a pull request in huggingface/transformers that received 7 comments
Add mBART-50
What does this PR do?
This is the second part of splitting #9811
This PR adds the mBART-50 models.
Add MBart50Tokenizer and MBart50TokenizerFast.…
+1,008
−34
•
7
comments
Opened 12 other pull requests in 1 repository
huggingface/transformers
3
open
9
merged
- TransCoder
- Add m2m100
- [WIP] SpeechToTextTransformer (ASR, ST)
- [RAG] fix tokenizer
- [WIP][examples/seq2seq] move old s2s scripts to legacy
- [examples/run_s2s] remove task_specific_params and update rouge computation
- [RAG] fix generate
- remove adjust_logits_during_generation method
- [examples/s2s] add test set predictions
- [run_clm.py] fix getting extention
- [docs] fix auto model docs
- Fix bart conversion script
Reviewed 14 pull requests in 1 repository
huggingface/transformers 14 pull requests
- Add m2m100
- set tgt_lang of MBart Tokenizer for summarization
- [run_glue] Add MNLI compatible mode
- Add mBART-50
- [RAG] fix tokenizer
- Fix typo in GPT2DoubleHeadsModel docs
- Update run_xnli.py to use Datasets library
- [examples/s2s] add test set predictions
- [s2s examples] Replace -100 token ids with the tokenizer pad_id for compute_metrics
- [s2s examples] README.md fixes
- [s2s] Can't mix --fp16 and --device cpu
- [examples/seq2seq] support label smoothing
- attention_mask -> encoder_attention_mask in cross attn of BERT-like models
- adapt mbart and generate for Mbart50
Created an issue in huggingface/transformers that received 1 comment
[mBART] one slow integration test is failing on master
The test_enro_generate_batch slow test is now failing on master
command
RUN_SLOW=1 pytest tests/test_modeling_mbart.py::MBartEnroIntegrationTest::t…
1
comment