Peter St. John bab40c6838 [core] support tensor-valued _extra_state values in from_pretrained (#38155) ...

Support tensor-valued _extra_state values

TransformerEngine uses the pytorch get/set_extra_state API to store FP8
layer config information as bytes Tensor in the _extra_state entry in
the state dict. With recent changes to from_pretrained, this
functionality has broken and loading a model that uses this API doesn't
appear to work. This PR fixes the save/load pretrained functions for
extra state entries that use a pytorch tensor, and adds a (currently
x-failing) test for a dictionary extra state.

Signed-off-by: Peter St. John <pstjohn@nvidia.com>

2025-05-28 15:38:42 +02:00

..

import_structures

Support for version spec in requires & arbitrary mismatching depths across folders (#37854)

2025-05-09 15:26:27 +02:00

__init__.py

[Test refactor 1/5] Per-folder tests reorganization (#15725)

2022-02-23 15:46:28 -05:00

test_activations_tf.py

TF: Add sigmoid activation function (#16819)

2022-04-19 16:13:08 +01:00

test_activations.py

use torch.testing.assertclose instead to get more details about error in cis (#35659)

2025-01-24 16:55:28 +01:00

test_add_new_model_like.py

Use Python 3.9 syntax in tests (#37343)

2025-04-08 14:12:08 +02:00

test_audio_utils.py

Use Python 3.9 syntax in tests (#37343)

2025-04-08 14:12:08 +02:00

test_auto_docstring.py

[AutoDocstring] Based on inspect parsing of the signature (#33771)

2025-05-08 17:46:07 -04:00

test_backbone_utils.py

🚨 out_indices always a list (#30941)

2024-05-22 15:23:04 +01:00

test_cache_utils.py

switch to device agnostic device calling for test cases (#38247)

2025-05-26 10:18:53 +02:00

test_chat_template_utils.py

Use Python 3.9 syntax in tests (#37343)

2025-04-08 14:12:08 +02:00

test_cli.py

Transformers cli clean command (#37657)

2025-04-30 12:15:43 +01:00

test_configuration_utils.py

🚨Early-error🚨 config will error out if output_attentions=True and the attn implementation is wrong (#38288)

2025-05-23 17:17:38 +02:00

test_convert_slow_tokenizer.py

Revert error back into warning for byte fallback conversion. (#22607)

2023-04-06 14:00:29 +02:00

test_deprecation.py

enable utils test cases on XPU (#38005)

2025-05-09 08:45:01 +02:00

test_doc_samples.py

Use Python 3.9 syntax in tests (#37343)

2025-04-08 14:12:08 +02:00

test_dynamic_module_utils.py

Fix the regex in get_imports to support multiline try blocks and excepts with specific exception types (#23725)

2023-05-24 15:40:19 -04:00

test_expectations.py

enable 2 llama UT cases on xpu (#37126)

2025-04-07 16:02:14 +02:00

test_feature_extraction_utils.py

Use Python 3.9 syntax in tests (#37343)

2025-04-08 14:12:08 +02:00

test_file_utils.py

Inheritance-based framework detection (#21784)

2023-02-27 15:31:55 +00:00

test_generic.py

Use Python 3.9 syntax in tests (#37343)

2025-04-08 14:12:08 +02:00

test_hf_argparser.py

Improve typing in TrainingArgument (#36944)

2025-05-21 13:54:38 +00:00

test_hub_utils.py

Add test to ensure unknown exceptions reraising in utils/hub.py::cached_files() (#37651)

2025-04-22 11:38:10 +02:00

test_image_processing_utils.py

Use Python 3.9 syntax in tests (#37343)

2025-04-08 14:12:08 +02:00

test_image_utils.py

🔴 Video processors as a separate class (#35206)

2025-05-12 11:55:51 +02:00

test_import_structure.py

Support for version spec in requires & arbitrary mismatching depths across folders (#37854)

2025-05-09 15:26:27 +02:00

test_import_utils.py

Fix test isolation for clear_import_cache utility (#36345)

2025-03-17 16:09:09 +01:00

test_logging.py

Fix flaky test for log level (#21776)

2023-02-28 16:24:14 -05:00

test_model_card.py

Use Python 3.9 syntax in tests (#37343)

2025-04-08 14:12:08 +02:00

test_model_debugging_utils.py

Model debugger upgrades (#37391)

2025-04-18 16:45:54 +02:00

test_model_output.py

Use Python 3.9 syntax in tests (#37343)

2025-04-08 14:12:08 +02:00

test_modeling_flax_utils.py

[tests] remove flax-pt equivalence and cross tests (#36283)

2025-02-19 15:13:27 +00:00

test_modeling_rope_utils.py

Use Python 3.9 syntax in tests (#37343)

2025-04-08 14:12:08 +02:00

test_modeling_tf_core.py

Use Python 3.9 syntax in tests (#37343)

2025-04-08 14:12:08 +02:00

test_modeling_tf_utils.py

Use Python 3.9 syntax in tests (#37343)

2025-04-08 14:12:08 +02:00

test_modeling_utils.py

[core] support tensor-valued _extra_state values in from_pretrained (#38155)

2025-05-28 15:38:42 +02:00

test_offline.py

Use Python 3.9 syntax in tests (#37343)

2025-04-08 14:12:08 +02:00

test_processing_utils.py

Use Python 3.9 syntax in tests (#37343)

2025-04-08 14:12:08 +02:00

test_skip_decorators.py

enable utils test cases on XPU (#38005)

2025-05-09 08:45:01 +02:00

test_tokenization_utils.py

Use Python 3.9 syntax in tests (#37343)

2025-04-08 14:12:08 +02:00

test_versions_utils.py

Remove old code for PyTorch, Accelerator and tokenizers (#37234)

2025-04-10 20:54:21 +02:00

test_video_utils.py

[video utils] group and reorder by number of frames (#38374)

2025-05-27 11:32:33 +02:00

tiny_model_summary.json

Add image text to text pipeline (#34170)

2024-10-31 15:48:11 -04:00