Skip to content

GitLab

"tests/vscode:/vscode.git/clone" did not exist on "93cd94b79d62a8a8b6f4f20e51fda95e6daa4d3a"
Switch branch/tag
  1. 23 Jun, 2022 3 commits
  3. 22 Jun, 2022 5 commits
    • Sylvain Gugger's avatar
      Offload fixes (#17810) · df8e6804
      Sylvain Gugger authored 
      * Offload fixes

* Add a test
      df8e6804
    • Joao Gante's avatar
      CLI: use hub's `create_commit` (#17755) · 0d0c392c
      Joao Gante authored 
      * use create_commit

* better commit message and description

* touch setup.py to trigger cache update

* add hub version gating
      0d0c392c
    • Arthur's avatar
      initial commit (#17818) · 56b83cf0
      Arthur authored
      56b83cf0
    • Eran Hirsch's avatar
      Add logits_processor parameter, used by `generate`, to `Seq2SeqTrainer`... · 13570381
      Eran Hirsch authored 
      Add logits_processor parameter, used by `generate`, to `Seq2SeqTrainer` methods `evaluate` and `predict` (#17805)

* Add logits_processor parameter, used by `generate`, to `Seq2SeqTrainer` methods `evaluate` and `predict`

* Add all generate parameters to `Seq2SeqTrainer`, and also to `QuestionAnsweringSeq2SeqTrainer` which overrides it

* Remove `self._num_beams` from trainer classes

* - Run fixup
- Fix "Constraint" not exposed
- Fix synced_gpus to actually read from param

* Use kwargs

* Copy kwargs before making changes to it

* Fix style issues unused imports
      13570381
    • Arthur's avatar
      Flax sharded (#17760) · 16c6eb7c
      Arthur authored
      16c6eb7c
  5. 21 Jun, 2022 13 commits
  7. 20 Jun, 2022 5 commits
  9. 17 Jun, 2022 3 commits
    • Swetha Mandava's avatar
      Save huggingface checkpoint as artifact in mlflow callback (#17686) · 522a9ece
      Swetha Mandava authored 
      

* Fix eval to compute rouge correctly for rouge_score

* styling

* moving sentence tokenization to utils from run_eval

* saving ckpt in mlflow

* use existing format of args

* fix documentation
Co-authored-by: default avatarSwetha Mandava <smandava@nvidia.com>
      522a9ece
    • Sourab Mangrulkar's avatar
      Migrate HFDeepSpeedConfig from trfrs to accelerate (#17623) · 21a77242
      Sourab Mangrulkar authored 
      

* Migrate HFDeepSpeedConfig from trfrs to accelerate

* add `accelerate` to testing dep

* addressing comments

* addressing comments

Using `_shared_state` and avoiding object creation. This is necessary as `notebook_launcher` in `launcers.py` checks `len(AcceleratorState._shared_state)>0` to throw an error.

* resolving comments

1. Use simple API from accelerate to manage the deepspeed config integration
2. Update the related documentation

* reverting changes and addressing comments

* docstring correction

* addressing nits

* addressing nits

* addressing nits 3

* bumping up the accelerate version to 0.10.0

* resolving import

* update setup.py to include deepspeed dependencies

* Update dependency_versions_table.py

* fixing imports

* reverting changes to CI dependencies for "run_tests_pipelines_tf*" tests

These changes didn't help with resolving the failures and I believe this needs to be addressed in another PR.

* removing `accelerate` as hard dependency

Resolves issues related to CI Tests

* adding `accelerate` as dependency for building docs

resolves failure in Build PR Documentation test

* adding `accelerate` as dependency in "dev" to resolve doc build issue

* resolving comments

1. adding `accelerate` to extras["all"]
2. Including check for accelerate too before import HFDeepSpeedConfig from there
Co-Authored-By: default avatarSylvain Gugger <35901082+sgugger@users.noreply.github.com>

* resolving comments
Co-authored-by: default avatarSylvain Gugger <35901082+sgugger@users.noreply.github.com>
      21a77242
    • greg2451's avatar
      feat: add num_workers arg to DataLoader (#17751) · 2d7c1bb1
      greg2451 authored
      2d7c1bb1
  11. 16 Jun, 2022 4 commits
  13. 15 Jun, 2022 3 commits
  15. 14 Jun, 2022 4 commits