Highlights
- Pro
Block or Report
Block or report patrickvonplaten
Report abuse
Contact GitHub support about this user’s behavior. Learn more about reporting abuse.
Report abusePopular repositories
-
Wav2Vec2_PyCTCDecode Public
Small repo describing how to use Hugging Face's Wav2Vec2 with PyCTCDecode
-
-
-
-
1,810 contributions in the last year
Contribution activity
May 2022
Created 33 commits in 7 repositories
Created 2 repositories
- patrickvonplaten/Megatron-LM Python
- patrickvonplaten/metaseq Python
Created a pull request in huggingface/transformers that received 12 comments
[Tests] Fix slow opt tests
Fixes failing circle ci tests:
tests/models/opt/test_modeling_opt.py::OPTEmbeddingsTest::test_logits
tests/models/opt/test_modeling_opt.py::OPTMode…
+4
−10
•
12
comments
Opened 15 other pull requests in 6 repositories
huggingface/transformers
8
merged
1
closed
- [Test OPT] Add batch generation test opt
- [OPT] Run test in lower precision on GPU
- [Generation] Fix Transition probs
- [Fix-copies] Correct main fix-copies
- [Test] Fix W2V-Conformer integration test
- [Tests] Fix opt integration test
- [T5] Fix init in TF and Flax for pretraining
- OPT - fix docstring and improve tests slighly
- Younes opt 350 m
sanchit-gandhi/seq2seq-speech
2
merged
huggingface/hub-docs
1
merged
patrickvonplaten/metaseq
1
merged
patrickvonplaten/Megatron-LM
1
merged
ngoyal2707/Megatron-LM
1
closed
Reviewed 37 pull requests in 7 repositories
huggingface/transformers
25 pull requests
- Adds support for OPT in Flax and TF.
- fix for 17292
- [OPT] Run test in lower precision on GPU
-
Use Accelerate in
from_pretrainedfor big model inference - docs for typical decoding
- Fix a TF-T5 test
- Fix -1e4 as attn mask
- Add support for pretraining recurring span selection to Splinter
- [T5] Fix init in TF and Flax for pretraining
- [LED] fix global_attention_mask not being passed for generation and docs clarification about grad checkpointing
- Add XLNet OnnxConfig
- [WIP][Speech Model] Add Emformer
- Improved Documentation for Encoder Decoder models
- Add Wav2Vec2Conformer
- [Tests] Fix slow opt tests
- Doctest longformer
- Fix Flava FlavaForPreTrainingIntegrationTest test
- OPTForCausalLM lm_head input size should be config.word_embed_proj_dim
- OPT-fix
- M-CTC-T Model
- OPT - fix docstring and improve tests slighly
- Add OPT
- [feat] Add FLAVA model
- Skip RoFormer ONNX test if rjieba not installed
- Deprecate model templates
- Some pull request reviews not shown.
sanchit-gandhi/seq2seq-speech
2 pull requests
huggingface/hub-docs
1 pull request
facebookresearch/metaseq
1 pull request
huggingface/accelerate
1 pull request
huggingface/huggingface_hub
1 pull request
huggingface/datasets
1 pull request
Created an issue in facebookresearch/metaseq that received 17 comments
[Community] OPT Inference in HF Transformers
You can now use the OPT models in Hugging Face Transformers Go here for details: https://twitter.com/huggingface/status/1524783489593360385 (Edited…
17
comments

