Skip to content

T5 kbqa#13

Closed
wangguojim wants to merge 3 commits into
masterfrom
t5_kbqa
Closed

T5 kbqa#13
wangguojim wants to merge 3 commits into
masterfrom
t5_kbqa

Conversation

@wangguojim
Copy link
Copy Markdown
Contributor

t5 ner task and generate qa pairs from knowledge graph triples

Comment thread examples/t5_kbqa/finetune_t5_ner.py Outdated

class SeqDataset(Dataset):
"""
针对特定数据集,定义一个相关的取数据的方式
Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

translate it into English?

Comment thread examples/t5_kbqa/finetune_t5_ner.py Outdated

if __name__ == '__main__':
data_name = 'kgclue'
train_path = '/mnt/T5_JIEBA/input/train.csv'
Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Could your give a demo datasets

@@ -0,0 +1,210 @@
import sys
sys.path.append('/data/wang/models/FlagAI')
Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

this is removable

@wangguojim
Copy link
Copy Markdown
Contributor Author

fixed

@marscrazy marscrazy closed this May 30, 2022
BAAI-WuDao added a commit that referenced this pull request Jun 29, 2022
* fixed tokenizer in glm blankfilling

Signed-off-by: Anhforth <yanzhaodong2021@163.com>

* Update glm_generate_samples_en.py

Co-authored-by: Anhforth <yanzhaodong2021@163.com>
Co-authored-by: Zac Liu <liuguang@baai.ac.cn>
@BAAI-OpenPlatform BAAI-OpenPlatform deleted the t5_kbqa branch February 1, 2023 04:26
ftgreat pushed a commit that referenced this pull request Mar 16, 2023
BAAI-OpenPlatform pushed a commit to FlagOpen/FlagAI that referenced this pull request Sep 20, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants