Merge pull request #82 from hughqiu/master

文学散文训练模型分享,同时添加模型共享列表
This commit is contained in:
Zeyao Du
2019-10-25 19:37:30 +08:00
committed by GitHub
4 changed files with 16 additions and 0 deletions

View File

@@ -71,6 +71,15 @@ python ./generate.py --length=50 --nsamples=4 --prefix=xxx --fast_pattern --save
howpublished = {\url{https://github.com/Morizeyao/GPT2-Chinese}},
}
```
## 模型分享
| 模型名称 | 模型介绍| 分享者| 链接地址1 | 链接地址2 |
| ------------ | ------------ | ------------ | ------------ | ------------ |
| 散文模型 | 使用130MB的名家散文、情感散文和散文诗歌训练所得 。 | [hughqiu](https://github.com/hughqiu "hughqiu") | [百度网盘【pp93】](https://pan.baidu.com/s/1C1i6lhxujsC_Swv-LSmWXw "百度网盘【pp93】") | [GDrive](https://drive.google.com/drive/folders/1rJC4niJKMVwixUQkuL9k5teLRnEYTmUf?usp=sharing "GDrive") |
此处为热情大方的git友训练所得的模型文件公开给所有朋友使用同时也欢迎各位伙伴将自己训练完毕的模型公开于此处。
## Demo
@@ -106,3 +115,10 @@ python ./generate.py --length=50 --nsamples=4 --prefix=xxx --fast_pattern --save
![avatar](sample/金庸_倚天屠龍記.jpg)
![avatar](sample/金庸_鹿鼎記.jpg)
![avatar](sample/金庸_神鵰俠侶.jpg)
-以下为文学散文的生成样例,由[hughqiu](https://github.com/hughqiu "hughqiu")贡献模型已经分享于模型分享列表。语料130MBBatch size 1610层深度下训练10轮所得。
![avatar](sample/散文1.png)
![avatar](sample/散文2.png)
![avatar](sample/散文3.png)

BIN
sample/散文1.png Normal file

Binary file not shown.

After

Width:  |  Height:  |  Size: 618 KiB

BIN
sample/散文2.png Normal file

Binary file not shown.

After

Width:  |  Height:  |  Size: 678 KiB

BIN
sample/散文3.png Normal file

Binary file not shown.

After

Width:  |  Height:  |  Size: 856 KiB