- 07 Dec, 2023 1 commit
-
-
Alex McKinney authored
* Copies `modeling_flax_gpt_neo.py` to start * MLP Block. WIP Attention and Block * Adds Flax implementation of `LlamaMLP` Validated with in-file test. Some slight numeric differences, but assuming it isn't an issue * Adds `FlaxLlamaRMSNorm` layer `flax.linen` includes `RMSNorm` layer but not necessarily in all versions. Hence, we add in-file. * Adds FlaxLlamaAttention Copied from GPT-J as it has efficient caching implementation as well as rotary embeddings. Notice numerically different, but not by a huge amount. Needs investigating * Adds `FlaxLlamaDecoderLayer` numerically inaccurate, debugging.. * debugging rotary mismatch gptj uses interleaved whilst llama uses contiguous i think they match now but still final result is wrong. maybe drop back to just debugging attention layer? * fixes bug with decoder layer still somewhat numerically inaccurate, but close enough for now * adds markers for what to implement next the structure here diverges a lot from the PT version. not a big fan of it, but just get something working for now * implements `FlaxLlamaBlockCollection`] tolerance must be higher than expected, kinda disconcerting * Adds `FlaxLlamaModule` equivalent PyTorch model is `LlamaModel` yay! a language model
馃 * adds `FlaxLlamaForCausalLMModule` equivalent to `LlamaForCausalLM` still missing returning dict or tuple, will add later * start porting pretrained wrappers realised it probably needs return dict as a prereq * cleanup, quality, style * readds `return_dict` and model output named tuples * (tentatively) pretrained wrappers work馃敟 * fixes numerical mismatch in `FlaxLlamaRMSNorm` seems `jax.lax.rsqrt` does not match `torch.sqrt`. manually computing `1 / jax.numpy.sqrt` results in matching values. * [WIP] debugging numerics * numerical match I think issue was accidental change of backend. forcing CPU fixes test. We expect some mismatch on GPU. * adds in model and integration tests for Flax Llama summary of failing: - mul invalid combination of dimensions - one numerical mismatch - bf16 conversion (maybe my local backend issue) - params are not FrozenDict * adds missing TYPE_CHECKING import and `make fixup` * adds back missing docstrings needs review on quality of docstrings, not sure what is required. Furthermore, need to check if `CHECKPOINT_FOR_DOC` is valid. See TODO * commenting out equivalence test as can just use common * debugging * Fixes bug where mask and pos_ids were swapped in pretrained models This results in all tests passing now馃敟 * cleanup of modeling file * cleanup of test file * Resolving simpler review comments * addresses more minor review comments * fixing introduced pytest errors from review * wip additional slow tests * wip tests need to grab a GPU machine to get real logits for comparison otherwise, slow tests should be okay * `make quality`, `make style` * adds slow integration tests - checking logits - checking hidden states - checking generation outputs * `make fix-copies` * fix mangled function following `make fix-copies` * adds missing type checking imports * fixes missing parameter checkpoint warning * more finegrained 'Copied from' tags avoids issue of overwriting `LLAMA_INPUTS_DOCSTRING` * swaps import guards ??? how did these get swapped initially? * removing `inv_freq` again as pytorch version has now removed * attempting to get CI to pass * adds doc entries for llama flax models * fixes typo in __init__.py imports * adds back special equivalence tests these come from the gpt neo flax tests. there is special behaviour for these models that needs to override the common version * overrides tests with dummy to see if CI passes need to fill in these tests later * adds my contribution to docs * `make style; make quality` * replaces random masking with fixed to work with flax version * `make quality; make style` * Update src/transformers/models/llama/modeling_flax_llama.py Co-authored-by:Sanchit Gandhi <93869735+sanchit-gandhi@users.noreply.github.com> * Update src/transformers/models/llama/modeling_flax_llama.py Co-authored-by:
-
- 31 Oct, 2023 1 commit
-
-
Hz, Ji authored
* device agnostic models testing * add decorator `require_torch_fp16` * make style * apply review suggestion * Oops, the fp16 decorator was misused
-
- 12 Oct, 2023 1 commit
-
-
Yih-Dar authored
* fix --------- Co-authored-by:ydshieh <ydshieh@users.noreply.github.com>
-
- 11 Oct, 2023 1 commit
-
-
Yih-Dar authored
* copied statement for test files --------- Co-authored-by:
-
- 02 Oct, 2023 1 commit
-
-
Arthur authored
* fix wav2vec2 * nit * stash * one more file to update * fix byt5 * vocab size is 256, don't change that! * use other revision * test persimon in smaller size * style * tests * nits * update add tokens from pretrained * test tokenization * nits * potential fnet fix? * more nits * nits * correct test * assert close * udpate * ouch * fix it * some more nits * FINALLU * use `adept` checkpoints * more adept checkpoints * that was invlved!
-
- 29 Sep, 2023 1 commit
-
-
Yih-Dar authored
skip Co-authored-by:
-
- 12 Sep, 2023 1 commit
-
-
Arthur authored
* intiial commit * updates * nits * update conversion script * update conversion script * use path to load * add tips etc * some modeling logic * modeling update * more nits * nits * normal layer norm * update config and doc * nits * update doc remove unused * update * fix inits and stuff * fixup * revert wrong changes * updates * more nits * add default config values to the configuration file * fixup happy * update * 2 tests left * update readmes * more nits * slow test and more documentation * update readme * fix licences * styling * use fast if possible when saving tokenizer * remove todo * remove tokenization tests * small last nits * Apply suggestions from code review Co-authored-by:
Matt <Rocketknight1@users.noreply.github.com> * nits to skip the timout doctest * fix integration test * fix test * update eos token * update to allow fast tokenization * styling * fix codeLlama as well for the update post processor * Apply suggestions from code review Co-authored-by:
amyeroberts <22614925+amyeroberts@users.noreply.github.com> * add more copied from statements * update * doc passes doctest * remove `# final layer norm?` * change docstring prompot * update * Update README.md Co-authored-by:
Sylvain Gugger <35901082+sgugger@users.noreply.github.com> * don't doctest the conversion script as it requires more packages * don't init a model in the config * oups * fix doctest --------- Co-authored-by:
-
