未验证 提交 bfa2652d 编写于 作者: N nbcc 提交者: GitHub

Merge pull request #658 from dingsiyu/ernie-doc-develop-new

add ernie-doc to ernie-develop
...@@ -122,6 +122,11 @@ pip install -e . ...@@ -122,6 +122,11 @@ pip install -e .
| [ERNIE Gen base for English](https://ernie-github.cdn.bcebos.com/model-ernie-gen-base-en.1.tar.gz) | L12H768A12 |ernie-gen-base-en| | [ERNIE Gen base for English](https://ernie-github.cdn.bcebos.com/model-ernie-gen-base-en.1.tar.gz) | L12H768A12 |ernie-gen-base-en|
| [ERNIE Gen Large for English](https://ernie-github.cdn.bcebos.com/model-ernie-gen-large-en.1.tar.gz)| L24H1024A16 | ernie-gen-large-en | | [ERNIE Gen Large for English](https://ernie-github.cdn.bcebos.com/model-ernie-gen-large-en.1.tar.gz)| L24H1024A16 | ernie-gen-large-en |
| [ERNIE Gen Large 430G for English](https://ernie-github.cdn.bcebos.com/model-ernie-gen-large-430g-en.1.tar.gz)| Layer:24, Hidden:1024, Heads:16 + 430G pretrain corpus | ernie-gen-large-430g-en | | [ERNIE Gen Large 430G for English](https://ernie-github.cdn.bcebos.com/model-ernie-gen-large-430g-en.1.tar.gz)| Layer:24, Hidden:1024, Heads:16 + 430G pretrain corpus | ernie-gen-large-430g-en |
| [ERNIE Doc Base for Chinese](https://ernie-github.cdn.bcebos.com/model-ernie-doc-base-zh.tar.gz)| L12H768A12 | ernie-doc-base-zh |
| [ERNIE Doc Base for English](https://ernie-github.cdn.bcebos.com/model-ernie-doc-base-en.tar.gz)| L12H768A12 | ernie-doc-base-en |
| [ERNIE Doc Large for English](https://ernie-github.cdn.bcebos.com/model-ernie-doc-large-en.tar.gz)| L24H1024A16 | ernie-doc-large-zh |
| [ERNIE Gram Base for Chinese](https://ernie-github.cdn.bcebos.com/model-ernie-gram-zh.1.tar.gz) | L12H768A12 | ernie-gram-zh |
| [ERNIE Gram Base for English](https://ernie-github.cdn.bcebos.com/model-ernie-gram-en.1.tar.gz) | L12H768A12 | ernie-gram-en |
##### 4. download datasets ##### 4. download datasets
......
...@@ -129,6 +129,11 @@ export PYTHONPATH=$PWD:$PYTHONPATH ...@@ -129,6 +129,11 @@ export PYTHONPATH=$PWD:$PYTHONPATH
| [ERNIE Gen Base 英文](https://ernie-github.cdn.bcebos.com/model-ernie-gen-base-en.1.tar.gz) | Layer:12, Hidden:768, Heads:12 |ernie-gen-base-en| | [ERNIE Gen Base 英文](https://ernie-github.cdn.bcebos.com/model-ernie-gen-base-en.1.tar.gz) | Layer:12, Hidden:768, Heads:12 |ernie-gen-base-en|
| [ERNIE Gen Large 英文](https://ernie-github.cdn.bcebos.com/model-ernie-gen-large-en.1.tar.gz)| Layer:24, Hidden:1024, Heads:16 |ernie-gen-large-en| | [ERNIE Gen Large 英文](https://ernie-github.cdn.bcebos.com/model-ernie-gen-large-en.1.tar.gz)| Layer:24, Hidden:1024, Heads:16 |ernie-gen-large-en|
| [ERNIE Gen Large 430G英文](https://ernie-github.cdn.bcebos.com/model-ernie-gen-large-430g-en.1.tar.gz)| Layer:24, Hidden:1024, Heads:16 + 额外430G 预训练语料 | ernie-gen-large-430g-en | | [ERNIE Gen Large 430G英文](https://ernie-github.cdn.bcebos.com/model-ernie-gen-large-430g-en.1.tar.gz)| Layer:24, Hidden:1024, Heads:16 + 额外430G 预训练语料 | ernie-gen-large-430g-en |
| [ERNIE Doc Base 中文](https://ernie-github.cdn.bcebos.com/model-ernie-doc-base-zh.tar.gz)| Layer:12, Hidden:768, Heads:12 |ernie-doc-base-zh|
| [ERNIE Doc Base 英文](https://ernie-github.cdn.bcebos.com/model-ernie-doc-base-en.tar.gz)| Layer:12, Hidden:768, Heads:12 |ernie-doc-base-en|
| [ERNIE Doc Large 英文](https://ernie-github.cdn.bcebos.com/model-ernie-doc-large-en.tar.gz)| Layer:24, Hidden:1024, Heads:16 |ernie-doc-large-en|
| [ERNIE Gram Base 中文](https://ernie-github.cdn.bcebos.com/model-ernie-gram-zh.1.tar.gz)| L12H768A12 |ernie-gram-zh|
| [ERNIE Gram Base 英文](https://ernie-github.cdn.bcebos.com/model-ernie-gram-en.1.tar.gz)| L12H768A12 |ernie-gram-en|
##### 4. 下载数据集 ##### 4. 下载数据集
......
```bash
_____ ____ _ _ ___ _____ ______ _____ _____
| ____| _ \| \ | |_ _| ____| | __ \/ ___ \/ ____|
| _| | |_) | \| || || _| _____| | \ | | | | /
| |___| _ <| |\ || || |__|_____| |__/ | |___| | \ ____
|_____|_| \_\_| \_|___|_____| |______/\ _____ /\ _____|
```
The `ERNIE-Doc` (including all our pre-trained models) has been released at [here](https://github.com/PaddlePaddle/ERNIE/tree/repro/ernie-doc).
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册