Pedram Rostami
commited on
Commit
•
9aad44b
1
Parent(s):
dd07a7b
Update README.md
Browse files
README.md
CHANGED
|
@@ -72,9 +72,8 @@ input_tokens = tokenizer(model_input, return_tensors="pt")
|
|
| 72 |
input_tokens = input_tokens.to(device)
|
| 73 |
generate_ids = model.generate(**input_tokens, max_new_tokens=512, do_sample=False, repetition_penalty=1.1)
|
| 74 |
model_output = tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
|
| 75 |
-
model_output = model_output.replace(model_input, "")
|
| 76 |
|
| 77 |
-
print(model_output)
|
| 78 |
```
|
| 79 |
|
| 80 |
### How to Quantize the Model
|
|
|
|
| 72 |
input_tokens = input_tokens.to(device)
|
| 73 |
generate_ids = model.generate(**input_tokens, max_new_tokens=512, do_sample=False, repetition_penalty=1.1)
|
| 74 |
model_output = tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
|
|
|
|
| 75 |
|
| 76 |
+
print(model_output[len(model_input) + 1: ])
|
| 77 |
```
|
| 78 |
|
| 79 |
### How to Quantize the Model
|