* remove xnli_compute_metrics, add load_dataset, load_metric, set_seed,metric.compute,load_metric
* fix
* fix
* fix
* push
* fix
* everything works
* fix init
* fix
* special treatment for sepconv1d
* style
* 🙏🏽
* add doc and cleanup
* fix doc
* fix doc again
* fix doc again
* Apply suggestions from code review
* make style
* Proposal that should work
* Remove needless code
* Fix test
* Apply suggestions from code review
* remove xnli_compute_metrics, add load_dataset, load_metric, set_seed,metric.compute,load_metric
* amend README
* removed data_args.task_name and replaced with task_name = "xnli"; use split function to load train and validation dataset separately; remove __post_init__; remove flag --task_name from README.
* removed dict task_to_keys, use str "xnli" instead of variable task_name, change preprocess_function to use examples["premise"], examples["hypothesis"] directly, remove sentence1_key and sentence2_key, change compute_metrics function to cater only to accuracy metric, add condition for train_langauge is None when using dataset.load_dataset()
* removed `torch.distributed.barrier()` and `import torch` as `from_pretrained` is able to do the work; amend README
* fix rag generate and tests
* put back adjust_logits_during_generation
* tests are okay
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>
* Unify logging with f-strings
* Get limits from MLflow rather than hardcode
* Add a check for parameter length overflow
Also constants are marked as internal
* Don't stop run in on_train_end
This causes bad behaviour when there is a seprarte validation step:
validation gets recorded as separate run.
* Fix style
* Cleaning up `ConversationalPipeline` to support more than DialoGPT.
Currently ConversationalPipeline was heavily biased towards DialoGPT
,which is the default model for this pipeline.
This PR proposes changes to put back the modifications specific to
DialoGPT into tokenizer-specific behavior wherever possible, by
creating `_build_conversation_input_ids` function that takes
conversation as input, and returns a list of ints corresponding
to the tokens. It feels natural to put here because all models
have probably different strategies to build input_ids from the
full conversation and it's the tokenizer's job to transform strings
into tokens (and vice-versa)
If `_build_conversation_input_ids` is missing, previous behavior is
used so we don't break anything so far (except for blenderbot where it's a fix).
This PR also contains a fix for too long inputs. There used
to be dead code for trying to limit the size of incoming input.
The introduced fixed is that we limit
within `_build_conversation_input_ids` to `tokenizer.model_max_length`.
It corresponds to the intent of the removed dead code and is actually
better because it corresponds to `model_max_length` which is different
from `max_length` (which is a default parameter for `generate`).
- Removed `history` logic from the Conversation as it's not relevant
anymore because tokenization logic has been moved to tokenizer.
And tokenizer cannot save any cache, and conversation cannot know
what is relevant or not.
Also it's not usable from `blenderbot` because the input_ids are
not append only (EOS tokens is always at the end).
- Added `iter_texts` method on `Conversation` because all
the code was literred with some form of this iteration of
past/generated_responses.
* Removing torch mention in types.
* Adding type checking to `_build_conversation_input_ids`.
* Fixing import in strings.
