Commit ·
48effef
1
Parent(s): 73ee34c
Update README.md
Browse files
README.md
CHANGED
|
@@ -19,7 +19,8 @@ It has been trained on a parallel corpus on several Chinese video games translat
|
|
| 19 |
|
| 20 |
Sample generation script :
|
| 21 |
|
| 22 |
-
```
|
|
|
|
| 23 |
tokenizer = transformers.AutoTokenizer.from_pretrained(r"path\to\checkpoint")
|
| 24 |
model = AutoModelForSeq2SeqLM.from_pretrained(r"path\to\checkpoint")
|
| 25 |
tokenizer.src_lang = "zh"
|
|
@@ -31,4 +32,6 @@ translated_tokens = model.generate(**inputs, num_beams=10, do_sample=True)
|
|
| 31 |
translation = tokenizer.batch_decode(translated_tokens, skip_special_tokens=True)[0]
|
| 32 |
|
| 33 |
print("CH : ", test_string , " // EN : ", translation)```
|
|
|
|
|
|
|
| 34 |
* Translation sample and comparison with Google Translate and DeepL : [Link to Spreadsheet](https://docs.google.com/spreadsheets/d/1J1i9P0nyI9q5-m2iZGSUatt3ZdHSxU8NOp9tJH7wxsk/edit?usp=sharing)
|
|
|
|
| 19 |
|
| 20 |
Sample generation script :
|
| 21 |
|
| 22 |
+
```python
|
| 23 |
+
from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
|
| 24 |
tokenizer = transformers.AutoTokenizer.from_pretrained(r"path\to\checkpoint")
|
| 25 |
model = AutoModelForSeq2SeqLM.from_pretrained(r"path\to\checkpoint")
|
| 26 |
tokenizer.src_lang = "zh"
|
|
|
|
| 32 |
translation = tokenizer.batch_decode(translated_tokens, skip_special_tokens=True)[0]
|
| 33 |
|
| 34 |
print("CH : ", test_string , " // EN : ", translation)```
|
| 35 |
+
```
|
| 36 |
+
|
| 37 |
* Translation sample and comparison with Google Translate and DeepL : [Link to Spreadsheet](https://docs.google.com/spreadsheets/d/1J1i9P0nyI9q5-m2iZGSUatt3ZdHSxU8NOp9tJH7wxsk/edit?usp=sharing)
|