Commit
•
b10bfb6
1
Parent(s):
1da63f7
Update README.md
Browse files
README.md
CHANGED
|
@@ -6,6 +6,22 @@ tags:
|
|
| 6 |
- gpt2
|
| 7 |
- vit
|
| 8 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 9 |
|
| 10 |
|
| 11 |
|
|
|
|
| 6 |
- gpt2
|
| 7 |
- vit
|
| 8 |
---
|
| 9 |
+
# 模型介绍
|
| 10 |
+
|
| 11 |
+
|
| 12 |
+
|
| 13 |
+
|
| 14 |
+
1. vit对图像做encoder，然后再用gpt2做decoder
|
| 15 |
+
2. vit模型使用的是`google/vit-base-patch16-224`, gpt2使用的是`yuanzhoulvpi/gpt2_chinese`
|
| 16 |
+
3. 本模型支持中文
|
| 17 |
+
|
| 18 |
+
|
| 19 |
+
# 训练代码
|
| 20 |
+
|
| 21 |
+
[https://github.com/yuanzhoulvpi2017/zero_nlp/tree/main/vit-gpt2-image-chinese-captioning](https://github.com/yuanzhoulvpi2017/zero_nlp/tree/main/vit-gpt2-image-chinese-captioning)
|
| 22 |
+
|
| 23 |
+
|
| 24 |
+
|
| 25 |
|
| 26 |
|
| 27 |
|