mirror of
https://github.com/huggingface/transformers.git
synced 2025-07-04 13:20:12 +06:00
7d7ceca396
* [WIP] Model card defaults * finetuned_from default value * Add all mappings to the mapping file * Be more defensive on finetuned_from arg * Add default task tag * Separate tags from tasks * Edge case for dataset * Apply suggestions from code review Co-authored-by: Lysandre Debut <lysandre@huggingface.co> Co-authored-by: Lysandre Debut <lysandre@huggingface.co>
107 lines
3.8 KiB
Python
107 lines
3.8 KiB
Python
# coding=utf-8
|
|
# Copyright 2020 The HuggingFace Inc. team.
|
|
#
|
|
# Licensed under the Apache License, Version 2.0 (the "License");
|
|
# you may not use this file except in compliance with the License.
|
|
# You may obtain a copy of the License at
|
|
#
|
|
# http://www.apache.org/licenses/LICENSE-2.0
|
|
#
|
|
# Unless required by applicable law or agreed to in writing, software
|
|
# distributed under the License is distributed on an "AS IS" BASIS,
|
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
# See the License for the specific language governing permissions and
|
|
# limitations under the License.
|
|
|
|
# this script remaps classes to class strings so that it's quick to load such maps and not require
|
|
# loading all possible modeling files
|
|
#
|
|
# it can be extended to auto-generate other dicts that are needed at runtime
|
|
|
|
|
|
import os
|
|
import sys
|
|
from os.path import abspath, dirname, join
|
|
|
|
|
|
git_repo_path = abspath(join(dirname(dirname(__file__)), "src"))
|
|
sys.path.insert(1, git_repo_path)
|
|
|
|
src = "src/transformers/models/auto/modeling_auto.py"
|
|
dst = "src/transformers/utils/modeling_auto_mapping.py"
|
|
|
|
|
|
if os.path.exists(dst) and os.path.getmtime(src) < os.path.getmtime(dst):
|
|
# speed things up by only running this script if the src is newer than dst
|
|
sys.exit(0)
|
|
|
|
# only load if needed
|
|
from transformers.models.auto.modeling_auto import ( # noqa
|
|
MODEL_FOR_CAUSAL_LM_MAPPING,
|
|
MODEL_FOR_IMAGE_CLASSIFICATION_MAPPING,
|
|
MODEL_FOR_MASKED_LM_MAPPING,
|
|
MODEL_FOR_MULTIPLE_CHOICE_MAPPING,
|
|
MODEL_FOR_NEXT_SENTENCE_PREDICTION_MAPPING,
|
|
MODEL_FOR_OBJECT_DETECTION_MAPPING,
|
|
MODEL_FOR_PRETRAINING_MAPPING,
|
|
MODEL_FOR_QUESTION_ANSWERING_MAPPING,
|
|
MODEL_FOR_SEQ_TO_SEQ_CAUSAL_LM_MAPPING,
|
|
MODEL_FOR_SEQUENCE_CLASSIFICATION_MAPPING,
|
|
MODEL_FOR_TABLE_QUESTION_ANSWERING_MAPPING,
|
|
MODEL_FOR_TOKEN_CLASSIFICATION_MAPPING,
|
|
MODEL_MAPPING,
|
|
MODEL_WITH_LM_HEAD_MAPPING,
|
|
)
|
|
|
|
|
|
# Those constants don't have a name attribute, so we need to define it manually
|
|
mappings = {
|
|
"MODEL_FOR_QUESTION_ANSWERING_MAPPING": MODEL_FOR_QUESTION_ANSWERING_MAPPING,
|
|
"MODEL_FOR_CAUSAL_LM_MAPPING": MODEL_FOR_CAUSAL_LM_MAPPING,
|
|
"MODEL_FOR_IMAGE_CLASSIFICATION_MAPPING": MODEL_FOR_IMAGE_CLASSIFICATION_MAPPING,
|
|
"MODEL_FOR_MASKED_LM_MAPPING": MODEL_FOR_MASKED_LM_MAPPING,
|
|
"MODEL_FOR_MULTIPLE_CHOICE_MAPPING": MODEL_FOR_MULTIPLE_CHOICE_MAPPING,
|
|
"MODEL_FOR_NEXT_SENTENCE_PREDICTION_MAPPING": MODEL_FOR_NEXT_SENTENCE_PREDICTION_MAPPING,
|
|
"MODEL_FOR_OBJECT_DETECTION_MAPPING": MODEL_FOR_OBJECT_DETECTION_MAPPING,
|
|
"MODEL_FOR_OBJECT_DETECTION_MAPPING": MODEL_FOR_OBJECT_DETECTION_MAPPING,
|
|
"MODEL_FOR_QUESTION_ANSWERING_MAPPING": MODEL_FOR_QUESTION_ANSWERING_MAPPING,
|
|
"MODEL_FOR_SEQ_TO_SEQ_CAUSAL_LM_MAPPING": MODEL_FOR_SEQ_TO_SEQ_CAUSAL_LM_MAPPING,
|
|
"MODEL_FOR_SEQUENCE_CLASSIFICATION_MAPPING": MODEL_FOR_SEQUENCE_CLASSIFICATION_MAPPING,
|
|
"MODEL_FOR_TABLE_QUESTION_ANSWERING_MAPPING": MODEL_FOR_TABLE_QUESTION_ANSWERING_MAPPING,
|
|
"MODEL_FOR_TOKEN_CLASSIFICATION_MAPPING": MODEL_FOR_TOKEN_CLASSIFICATION_MAPPING,
|
|
"MODEL_MAPPING": MODEL_MAPPING,
|
|
"MODEL_WITH_LM_HEAD_MAPPING": MODEL_WITH_LM_HEAD_MAPPING,
|
|
}
|
|
|
|
|
|
def get_name(value):
|
|
if isinstance(value, tuple):
|
|
return tuple(get_name(o) for o in value)
|
|
return value.__name__
|
|
|
|
|
|
content = [
|
|
"# THIS FILE HAS BEEN AUTOGENERATED. To update:",
|
|
"# 1. modify: models/auto/modeling_auto.py",
|
|
"# 2. run: python utils/class_mapping_update.py",
|
|
"from collections import OrderedDict",
|
|
"",
|
|
]
|
|
|
|
for name, mapping in mappings.items():
|
|
entries = "\n".join([f' ("{k.__name__}", "{get_name(v)}"),' for k, v in mapping.items()])
|
|
|
|
content += [
|
|
"",
|
|
f"{name}_NAMES = OrderedDict(",
|
|
" [",
|
|
entries,
|
|
" ]",
|
|
")",
|
|
"",
|
|
]
|
|
|
|
print(f"Updating {dst}")
|
|
with open(dst, "w", encoding="utf-8", newline="\n") as f:
|
|
f.write("\n".join(content))
|