* rembert: fix python codeblock
* rembert: use correct google/rembert checkpoint name in documentation
* rembert: use correct google/rembert checkpoint name in TF documentation
* add new bloom classes
* (feat) add bloom classification tests; make style
* style: change import in test
* add some typehints to bloom classes
* merge main into branch
* fix: input checking in bloom seq classification
* fix tests
* change model class tests
* fix few tests
- more tests should pass
- one test left
* make token classifier return hidden states
* style: make BLOOM typehints consistent
Co-authored-by: Younes Belkada <49240599+younesbelkada@users.noreply.github.com>
Co-authored-by: younesbelkada <younesbelkada@gmail.com>
Co-authored-by: Younes Belkada <49240599+younesbelkada@users.noreply.github.com>
* Initial commit
* Make some fixes
* Make PT model full forward pass
* Drop TF & Flax implementation, fix copies etc
* Add Flax model and update some corresponding stuff
* Drop some TF things
* Update config and flax local attn
* Add encoder_attention_type to config
* .
* Update docs
* Do some cleansing
* Fix some issues -> make style; add some docs
* Fix position_bias + mask addition + Update tests
* Fix repo consistency
* Fix model consistency by removing flax operation over attn_mask
* [WIP] Add PT TGlobal LongT5
* .
* [WIP] Add flax tglobal model
* [WIP] Update flax model to use the right attention type in the encoder
* Fix flax tglobal model forward pass
* Make the use of global_relative_attention_bias
* Add test suites for TGlobal model
* Fix minor bugs, clean code
* Fix pt-flax equivalence though not convinced with correctness
* Fix LocalAttn implementation to match the original impl. + update READMEs
* Few updates
* Update: [Flax] improve large model init and loading #16148
* Add ckpt conversion script accoring to #16853 + handle torch device placement
* Minor updates to conversion script.
* Typo: AutoModelForSeq2SeqLM -> FlaxAutoModelForSeq2SeqLM
* gpu support + dtype fix
* Apply some suggestions from code review
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>
* * Remove (de)parallelize stuff
* Edit shape comments
* Update README.md
* make fix-copies
* Remove caching logic for local & tglobal attention
* Apply another batch of suggestions from code review
* Add missing checkpoints
* Format converting scripts
* Drop (de)parallelize links from longT5 mdx
* Fix converting script + revert config file change
* Revert "Remove caching logic for local & tglobal attention"
This reverts commit 2a619828f6ddc3e65bd9bb1725a12b77fa883a46.
* Stash caching logic in Flax model
* Make side relative bias used always
* Drop caching logic in PT model
* Return side bias as it was
* Drop all remaining model parallel logic
* Remove clamp statements
* Move test files to the proper place
* Update docs with new version of hf-doc-builder
* Fix test imports
* Make some minor improvements
* Add missing checkpoints to docs
* Make TGlobal model compatible with torch.onnx.export
* Replace some np.ndarray with jnp.ndarray
* Fix TGlobal for ONNX conversion + update docs
* fix _make_global_fixed_block_ids and masked neg value
* update flax model
* style and quality
* fix imports
* remove load_tf_weights_in_longt5 from init and fix copies
* add slow test for TGlobal model
* typo fix
* Drop obsolete is_parallelizable and one warning
* Update __init__ files to fix repo-consistency
* fix pipeline test
* Fix some device placements
* [wip]: Update tests -- need to generate summaries to update expected_summary
* Fix quality
* Update LongT5 model card
* Update (slow) summarization tests
* make style
* rename checkpoitns
* finish
* fix flax tests
Co-authored-by: phungvanduy <pvduy23@gmail.com>
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>
Co-authored-by: patil-suraj <surajp815@gmail.com>
* enable cpu distribution training using mpirun
*command like
* mpirun -n 2 python3 run_qa.py --no_cuda --xpu_backend ccl xxxx
*MASTER_ADDR and MASTER_PORT should be set as env
*export MASTER_ADDR=127.0.0.1
*export MASTER_PORT=29500
Signed-off-by: Wang, Yi A <yi.a.wang@intel.com>
* fix according to the review comment
Signed-off-by: Wang, Yi A <yi.a.wang@intel.com>
* use accelerate logic for cpu distribution training to set "RANK","LOCAL_RANK","WORLD_SIZE" environment
Signed-off-by: Wang, Yi A <yi.a.wang@intel.com>
* allow scope from trainer arg
* add ray_scope to training args
* escape double quotes
* make style && quality
* attempt to solve doc style issues
* splitting up URLs for style
* make fixup
* Update src/transformers/training_args.py
Co-authored-by: Antoni Baum <antoni.baum@protonmail.com>
* make style
Co-authored-by: Antoni Baum <antoni.baum@protonmail.com>
I'm guessing that the intention was to have the `_no_split_modules` class attribute for `GPTNeoXPreTrainedModel` to be set to `["GPTNeoXLayer"]`, akin to how its set as `["GPTJBlock"]` for `GPTJPreTrainedModel`.
If this is incorrect, please feel free to just close the PR.
Thanks!
* Raise RepoNotFoundError in case of 401
* Include changes from revert-17646-skip_repo_not_found
* Add a comment
* 💄 Code quality
* 💚 Update `get_from_cache` test
* 💚 Code quality & skip failing test
VisibleDeprecationWarning is addressed by specifying dtype=object when creating numpy array.
Update code based on review feedback.
Undo whitespace changes to tokenization_utils_base.py.
Co-authored-by: I like data <ilikedata@nym.hush.com>
When we're preparing the tensors for CPU for postprocessing, we need
to upgrade the `float16` to `float32` since CPUs don't have instructions
for `[b]float16`.
