Add byt5 Model #1742
Open
Add byt5 Model #1742
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
|
yingyibiao
reviewed
Apr 14, 2022
paddlenlp/datasets/__init__.py
Outdated
Comment on lines
46
to
47
| from .tweetqa import * | ||
| from .xsum import * |
There was a problem hiding this comment.
我们目前重新设计了我们的 datasets，实现了对 HF datasets 的全面兼容，使用方法同 HF 接口，所以目前不需要在代码中新增数据集相关文件。
paddlenlp/datasets/tweetqa.py
Outdated
Comment on lines
1
to
56
| # Copyright (c) 2020 PaddlePaddle Authors. All Rights Reserved. | ||
| # | ||
| # 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. | ||
|
|
||
| import collections | ||
| import json | ||
| import os | ||
|
|
||
| from paddle.dataset.common import md5file | ||
| from paddle.utils.download import get_path_from_url | ||
| from paddlenlp.utils.env import DATA_HOME | ||
| from . import DatasetBuilder | ||
|
|
||
| __all__ = ['tweetqa'] | ||
|
|
||
|
|
||
| class tweetqa(DatasetBuilder): | ||
|
|
||
| META_INFO = collections.namedtuple('META_INFO', ('file', 'md5', 'URL')) | ||
| SPLITS = { | ||
| 'train': META_INFO( | ||
| os.path.join('train.json'), '4e06fd1cfd5e7f0380499df8cbe17237', | ||
| 'https://bj.bcebos.com/paddlenlp/datasets/tweetqa/train.json'), | ||
| 'dev': META_INFO( | ||
| os.path.join('dev.json'), '9c39d49d25d5296bdc537409208ddc85', | ||
| 'https://bj.bcebos.com/paddlenlp/datasets/tweetqa/dev.json'), | ||
| 'test': META_INFO( | ||
| os.path.join('test.json'), '9c39d49d25d5296bdc537409208ddc85', | ||
| 'https://bj.bcebos.com/paddlenlp/datasets/tweetqa/test.json') | ||
| } | ||
|
|
||
| def _get_data(self, mode, **kwargs): | ||
| default_root = os.path.join(DATA_HOME, self.__class__.__name__) | ||
| filename, data_hash, URL = self.SPLITS[mode] | ||
| fullname = os.path.join(default_root, filename) | ||
| if not os.path.exists(fullname) or (data_hash and | ||
| not md5file(fullname) == data_hash): | ||
| get_path_from_url(URL, default_root) | ||
| return fullname | ||
|
|
||
| def _read(self, filename, split): | ||
| with open(filename, encoding="utf-8") as f: | ||
| tweet_qa = json.load(f) | ||
| for data in tweet_qa: | ||
| yield {"Question": data["Question"], "label": [] if split == "test" else data["Answer"],"Tweet": data["Tweet"],"qid": data["qid"]} | ||
|
|
paddlenlp/datasets/xsum.py
Outdated
Comment on lines
1
to
55
| # Copyright (c) 2020 PaddlePaddle Authors. All Rights Reserved. | ||
| # | ||
| # 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. | ||
|
|
||
| import collections | ||
| import json | ||
| import os | ||
|
|
||
| from paddle.dataset.common import md5file | ||
| from paddle.utils.download import get_path_from_url | ||
| from paddlenlp.utils.env import DATA_HOME | ||
| from . import DatasetBuilder | ||
|
|
||
| __all__ = ['xsum'] | ||
|
|
||
|
|
||
| class xsum(DatasetBuilder): | ||
|
|
||
| META_INFO = collections.namedtuple('META_INFO', ('file', 'md5', 'URL')) | ||
| SPLITS = { | ||
| 'train': META_INFO( | ||
| os.path.join('train.json'), '4e06fd1cfd5e7f0380499df8cbe17237', | ||
| 'https://bj.bcebos.com/paddlenlp/datasets/xsum/train.json'), | ||
| 'dev': META_INFO( | ||
| os.path.join('dev.json'), '9c39d49d25d5296bdc537409208ddc85', | ||
| 'https://bj.bcebos.com/paddlenlp/datasets/xsum/dev.json'), | ||
| 'test': META_INFO( | ||
| os.path.join('test.json'), '9c39d49d25d5296bdc537409208ddc85', | ||
| 'https://bj.bcebos.com/paddlenlp/datasets/xsum/test.json') | ||
| } | ||
|
|
||
| def _get_data(self, mode, **kwargs): | ||
| default_root = os.path.join(DATA_HOME, self.__class__.__name__) | ||
| filename, data_hash, URL = self.SPLITS[mode] | ||
| fullname = os.path.join(default_root, filename) | ||
| if not os.path.exists(fullname) or (data_hash and | ||
| not md5file(fullname) == data_hash): | ||
| get_path_from_url(URL, default_root) | ||
| return fullname | ||
|
|
||
| def _read(self, filename, split): | ||
| with open(filename, encoding="utf-8") as f: | ||
| xsums = json.load(f) | ||
| for data in xsums: | ||
| yield {"document": data["document"], "label": data["summary"],"id": data["id"]} |
yingyibiao
reviewed
Apr 14, 2022
| @@ -0,0 +1,383 @@ | |||
| # Copyright (c) 2021 PaddlePaddle Authors. All Rights Reserved. | |||
|
This Pull Request is stale because it has been open for 60 days with no activity. 当前Pull Request 60天内无活动，被标记为stale。 |
|
This Pull Request is stale because it has been open for 60 days with no activity. 当前Pull Request 60天内无活动，被标记为stale。 |
|
This Pull Request is stale because it has been open for 60 days with no activity. 当前Pull Request 60天内无活动，被标记为stale。 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
6 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Add model and datatsets
PR changes
Add new model byt5; Add two datasets tweetqa and xsum
Description
The link of the model byt5-small is https://aistudio.baidu.com/aistudio/datasetdetail/123125 ; The link of the datatset xsum is https://aistudio.baidu.com/aistudio/datasetdetail/122619 ; The link of the dataset tweetqa is https://aistudio.baidu.com/aistudio/datasetdetail/131440 ; Please upload the model and datasets and I will update the link and md5 of these datasets in my code