Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,45 @@
|
|
1 |
-
---
|
2 |
-
license: mit
|
3 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: mit
|
3 |
+
---
|
4 |
+
|
5 |
+
This is a new kind of model optimization.
|
6 |
+
|
7 |
+
A paper is currently being written on the technique.
|
8 |
+
|
9 |
+
## Quickstart
|
10 |
+
|
11 |
+
Here provides a code snippet with `apply_chat_template` to show you how to load the tokenizer and model and how to generate contents.
|
12 |
+
|
13 |
+
```python
|
14 |
+
from transformers import AutoModelForCausalLM, AutoTokenizer
|
15 |
+
device = "cuda" # the device to load the model onto
|
16 |
+
|
17 |
+
model = AutoModelForCausalLM.from_pretrained(
|
18 |
+
"dnhkng/RYS-XLarge",
|
19 |
+
torch_dtype="auto",
|
20 |
+
device_map="auto"
|
21 |
+
)
|
22 |
+
tokenizer = AutoTokenizer.from_pretrained("dnhkng/RYS-XLarge")
|
23 |
+
|
24 |
+
prompt = "Give me a short introduction to large language model."
|
25 |
+
messages = [
|
26 |
+
{"role": "system", "content": "You are a helpful assistant."},
|
27 |
+
{"role": "user", "content": prompt}
|
28 |
+
]
|
29 |
+
text = tokenizer.apply_chat_template(
|
30 |
+
messages,
|
31 |
+
tokenize=False,
|
32 |
+
add_generation_prompt=True
|
33 |
+
)
|
34 |
+
model_inputs = tokenizer([text], return_tensors="pt").to(device)
|
35 |
+
|
36 |
+
generated_ids = model.generate(
|
37 |
+
model_inputs.input_ids,
|
38 |
+
max_new_tokens=512
|
39 |
+
)
|
40 |
+
generated_ids = [
|
41 |
+
output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
|
42 |
+
]
|
43 |
+
|
44 |
+
response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
|
45 |
+
```
|