transformers/src/transformers
Yih-Dar 8635407bc7
Fix tf.concatenate + test past_key_values for TF models (#15774)
* fix wrong method name tf.concatenate

* add tests related to causal LM / decoder

* make style and quality

* clean-up

* Fix TFBertModel's extended_attention_mask when past_key_values is provided

* Fix tests

* fix copies

* More tf.int8 -> tf.int32 in TF test template

* clean-up

* Update TF test template

* revert the previous commit + update the TF test template

* Fix TF template extended_attention_mask when past_key_values is provided

* Fix some styles manually

* clean-up

* Fix ValueError: too many values to unpack in the test

* Fix more: too many values to unpack in the test

* Add a comment for extended_attention_mask when there is past_key_values

* Fix TFElectra extended_attention_mask when past_key_values is provided

* Add tests to other TF models

* Fix for TF Electra test: add prepare_config_and_inputs_for_decoder

* Fix not passing training arg to lm_head in TFRobertaForCausalLM

* Fix tests (with past) for TF Roberta

* add testing for pask_key_values for TFElectra model

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
2022-02-25 17:11:46 +01:00
..
benchmark Fix missing tpu variable in benchmark_args_tf.py (#13968) 2021-10-11 23:30:03 -04:00
commands Fix add-new-model-like when old model checkpoint is not found (#15805) 2022-02-24 08:58:18 +01:00
data Fixes tf_default_data_collator sometimes guessing the wrong dtype for labels (#15234) 2022-01-20 14:26:51 +00:00
models Fix tf.concatenate + test past_key_values for TF models (#15774) 2022-02-25 17:11:46 +01:00
onnx Enable ONNX export when PyTorch and TensorFlow installed in the same environment (#15625) 2022-02-11 16:25:06 +01:00
pipelines Adding the option to return_timestamps on pure CTC ASR models. (#15792) 2022-02-25 14:06:45 +01:00
sagemaker Removes SageMakerTrainer code but keeps class as wrapper (#11587) 2021-05-04 14:31:18 -04:00
utils HFTracer.trace should use/return self.graph to be compatible with torch.fx.Tracer (#15824) 2022-02-25 15:54:45 +01:00
__init__.py Enable image-segmentation on AutoModelForSemanticSegmentation (#15647) 2022-02-23 17:20:26 +01:00
activations_tf.py Gelu10 (#15676) 2022-02-22 18:21:16 +01:00
activations.py Gelu10 (#15676) 2022-02-22 18:21:16 +01:00
configuration_utils.py Remove redundant error logging in from_pretrained() method (#15631) 2022-02-14 18:03:07 +01:00
convert_graph_to_onnx.py Prepare deprecated ONNX exporter for torch v1.11 (#15388) 2022-01-28 16:32:47 +01:00
convert_pytorch_checkpoint_to_tf2.py Doc styler v2 (#14950) 2021-12-27 16:31:21 -05:00
convert_slow_tokenizer.py fix CLIP fast tokenizer and change some properties of the slow version (#15067) 2022-02-18 10:21:30 +01:00
convert_slow_tokenizers_checkpoints_to_fast.py Doc styler v2 (#14950) 2021-12-27 16:31:21 -05:00
convert_tf_hub_seq_to_seq_bert_to_pytorch.py Enforce string-formatting with f-strings (#10980) 2021-03-31 10:00:27 -04:00
debug_utils.py Doc styler examples (#14953) 2021-12-27 19:07:46 -05:00
deepspeed.py Doc styler v2 (#14950) 2021-12-27 16:31:21 -05:00
dependency_versions_check.py [setup] make fairscale and deepspeed setup extras (#11151) 2021-04-08 15:46:54 -07:00
dependency_versions_table.py fix CLIP fast tokenizer and change some properties of the slow version (#15067) 2022-02-18 10:21:30 +01:00
dynamic_module_utils.py Allow custom code for Processors (#15649) 2022-02-15 09:44:35 -05:00
feature_extraction_sequence_utils.py [doc] :class: hunt (#14955) 2021-12-27 17:17:38 -08:00
feature_extraction_utils.py Remove redundant error logging in from_pretrained() method (#15631) 2022-02-14 18:03:07 +01:00
file_utils.py fix CLIP fast tokenizer and change some properties of the slow version (#15067) 2022-02-18 10:21:30 +01:00
generation_beam_constraints.py Constrained Beam Search [without disjunctive decoding] (#15416) 2022-02-09 16:59:26 +01:00
generation_beam_search.py Constrained Beam Search [without disjunctive decoding] (#15416) 2022-02-09 16:59:26 +01:00
generation_flax_logits_process.py TF generate refactor - Greedy Search (#15562) 2022-02-15 17:54:43 +01:00
generation_flax_utils.py Upgrade black to version ~=22.0 (#15565) 2022-02-09 09:28:57 -05:00
generation_logits_process.py TF generate refactor - Greedy Search (#15562) 2022-02-15 17:54:43 +01:00
generation_stopping_criteria.py Doc styler v2 (#14950) 2021-12-27 16:31:21 -05:00
generation_tf_logits_process.py TF generate refactor - Greedy Search (#15562) 2022-02-15 17:54:43 +01:00
generation_tf_utils.py TF generate refactor - Greedy Search (#15562) 2022-02-15 17:54:43 +01:00
generation_utils.py Update bad_words_ids usage (#15641) 2022-02-15 16:44:34 +01:00
hf_argparser.py Fix HfArgumentParser when passing a generator (#15758) 2022-02-23 00:16:38 +01:00
image_utils.py Add option to resize like torchvision's Resize (#15419) 2022-02-02 09:44:22 +01:00
integrations.py Add W&B backend for hyperparameter sweep (#14582) 2022-02-02 14:06:14 -05:00
keras_callbacks.py Example script for PushToHubCallback (#15375) 2022-01-27 16:16:24 +00:00
modelcard.py Fix a typo in tag addition (#15286) 2022-01-24 07:21:42 -05:00
modeling_flax_outputs.py Doc styler v2 (#14950) 2021-12-27 16:31:21 -05:00
modeling_flax_pytorch_utils.py Revert "Handle PyTorch to Flax conversion of 1D convolutions (#15519)" (#15540) 2022-02-07 12:33:49 +01:00
modeling_flax_utils.py Remove redundant error logging in from_pretrained() method (#15631) 2022-02-14 18:03:07 +01:00
modeling_outputs.py Revert changes in logit size for semantic segmentation models (#15722) 2022-02-24 15:52:52 +01:00
modeling_tf_outputs.py Doc styler v2 (#14950) 2021-12-27 16:31:21 -05:00
modeling_tf_pytorch_utils.py Add TFSpeech2Text (#15113) 2022-02-08 16:27:23 +00:00
modeling_tf_utils.py TF train_step docstring (#15755) 2022-02-22 11:18:35 +00:00
modeling_utils.py Fix from_pretrained with default base_model_prefix (#15814) 2022-02-24 11:43:51 +01:00
optimization_tf.py [doc] consistent True/False/None default format (#14951) 2021-12-27 14:31:40 -08:00
optimization.py Add a missing space in a deprecation message (#15651) 2022-02-15 19:12:30 -05:00
processing_utils.py Add push_to_hub method to processors (#15668) 2022-02-15 21:14:04 -05:00
py.typed Create py.typed (#12893) 2021-08-13 04:12:59 -04:00
pytorch_utils.py [torch_int_div] Correct true division in generation (#15498) 2022-02-07 16:04:18 +01:00
testing_utils.py [Test refactor 1/5] Per-folder tests reorganization (#15725) 2022-02-23 15:46:28 -05:00
tf_utils.py TF generate refactor - Greedy Search (#15562) 2022-02-15 17:54:43 +01:00
tokenization_utils_base.py Remove redundant error logging in from_pretrained() method (#15631) 2022-02-14 18:03:07 +01:00
tokenization_utils_fast.py fix the tokenizer_config.json file for the slow tokenizer when a fast version is available (#15319) 2022-02-01 16:48:25 +01:00
tokenization_utils.py Fixing a pathological case for slow tokenizers (#14981) 2021-12-30 09:10:34 +01:00
trainer_callback.py [Trainer] Deeper length checks for IterableDatasetShard (#15539) 2022-02-07 18:34:56 +03:00
trainer_pt_utils.py Doc styler examples (#14953) 2021-12-27 19:07:46 -05:00
trainer_seq2seq.py Fix Seq2SeqTrainer (#15603) 2022-02-10 16:26:14 +01:00
trainer_tf.py Doc styler v2 (#14950) 2021-12-27 16:31:21 -05:00
trainer_utils.py [Trainer] Deeper length checks for IterableDatasetShard (#15539) 2022-02-07 18:34:56 +03:00
trainer.py fix: hfdeepspeed config argument (#15711) 2022-02-18 12:00:02 -05:00
training_args_seq2seq.py Doc styler v2 (#14950) 2021-12-27 16:31:21 -05:00
training_args_tf.py Doc styler v2 (#14950) 2021-12-27 16:31:21 -05:00
training_args.py Set syncfree AdamW as the default optimizer for xla:gpu device in amp mode (#15361) 2022-01-27 20:05:31 -05:00