Update app.py
Browse files
app.py
CHANGED
@@ -1,3 +1,31 @@
|
|
1 |
-
|
2 |
-
|
3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import torch
|
2 |
+
from transformers import AutoModelForCausalLM, AutoTokenizer
|
3 |
+
|
4 |
+
model_id = "edumunozsala/llama-2-7b-int4-python-code-20k"
|
5 |
+
|
6 |
+
tokenizer = AutoTokenizer.from_pretrained(hf_model_repo)
|
7 |
+
|
8 |
+
model = AutoModelForCausalLM.from_pretrained(hf_model_repo, load_in_4bit=True, torch_dtype=torch.float16,
|
9 |
+
device_map=device_map)
|
10 |
+
|
11 |
+
instruction="Write a Python function to display the first and last elements of a list."
|
12 |
+
input=""
|
13 |
+
|
14 |
+
prompt = f"""### Instruction:
|
15 |
+
Use the Task below and the Input given to write the Response, which is a programming code that can solve the Task.
|
16 |
+
|
17 |
+
### Task:
|
18 |
+
{instruction}
|
19 |
+
|
20 |
+
### Input:
|
21 |
+
{input}
|
22 |
+
|
23 |
+
### Response:
|
24 |
+
"""
|
25 |
+
|
26 |
+
input_ids = tokenizer(prompt, return_tensors="pt", truncation=True).input_ids.cuda()
|
27 |
+
# with torch.inference_mode():
|
28 |
+
outputs = model.generate(input_ids=input_ids, max_new_tokens=100, do_sample=True, top_p=0.9,temperature=0.5)
|
29 |
+
|
30 |
+
print(f"Prompt:\n{prompt}\n")
|
31 |
+
print(f"Generated instruction:\n{tokenizer.batch_decode(outputs.detach().cpu().numpy(), skip_special_tokens=True)[0][len(prompt):]}")
|