Update README.md
Browse files
README.md
CHANGED
@@ -9,6 +9,17 @@ widget:
|
|
9 |
As we all know, the single direction language model based on decoder structure has strong generation ability, such as GPT model. **The 3.5 billion parameter Wenzhong-3.5B large model, using 100G chinese common data, 32 A100 training for 28 hours,** is the largest open source **GPT2 large model of chinese**. **Our model performs well in Chinese continuation generation.**
|
10 |
|
11 |
## Usage
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
12 |
```python
|
13 |
from transformers import pipeline, set_seed
|
14 |
|
|
|
9 |
As we all know, the single direction language model based on decoder structure has strong generation ability, such as GPT model. **The 3.5 billion parameter Wenzhong-3.5B large model, using 100G chinese common data, 32 A100 training for 28 hours,** is the largest open source **GPT2 large model of chinese**. **Our model performs well in Chinese continuation generation.**
|
10 |
|
11 |
## Usage
|
12 |
+
|
13 |
+
### load model
|
14 |
+
```python
|
15 |
+
from transformers import GPT2Tokenizer, GPT2Model
|
16 |
+
tokenizer = GPT2Tokenizer.from_pretrained('IDEA-CCNL/Wenzhong-3.5B')
|
17 |
+
model = GPT2Model.from_pretrained('IDEA-CCNL/Wenzhong-3.5B')
|
18 |
+
text = "Replace me by any text you'd like."
|
19 |
+
encoded_input = tokenizer(text, return_tensors='pt')
|
20 |
+
output = model(**encoded_input)
|
21 |
+
```
|
22 |
+
### generation
|
23 |
```python
|
24 |
from transformers import pipeline, set_seed
|
25 |
|