Are you sure you want to delete this task? Once this task is deleted, it cannot be recovered.
You can not select more than 25 topics
Topics must start with a chinese character,a letter or number, can include dashes ('-') and can be up to 35 characters long.
GQY
ae1a3eff00
|
1 year ago | |
|---|---|---|
| README.md | 1 year ago | |
| Untitled.ipynb | 1 year ago | |
| Untitled1.ipynb | 1 year ago | |
| WebServing_rofomers.py | 1 year ago | |
| base_serving.py | 1 year ago | |
| base_serving_faster.py | 1 year ago | |
| clean_data.ipynb | 1 year ago | |
| load_test.ipynb | 1 year ago | |
| load_test.py | 1 year ago | |
| predict_test-ml512.py | 1 year ago | |
| predict_test.py | 1 year ago | |
| roformer_v2-L-24_H-1024_A-16.ipynb | 1 year ago | |
| roformer_v2.ipynb | 1 year ago | |
| single.py | 1 year ago | |
| untitled.py | 1 year ago | |
README.md
本项目是fork追一科技的rofomers v2版本与bert4keras example中的单机多卡训练unilm文本生成模型
数据及来源为互联网科学问答数据集
预训练语言模型为追一科技开源chinese—roformer-v2 L12预训练语言模型与L24预训练语言模型
single中为单机单卡训练
untitled中为单机多卡训练
base_serving 为基于gradio部署tensorflow、keras模型。