Spaces:
Runtime error
Runtime error
Upload folder using huggingface_hub
Browse files
app.py
CHANGED
@@ -3,13 +3,8 @@ from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline
|
|
3 |
import torch
|
4 |
import gradio as gr
|
5 |
import spaces
|
6 |
-
from dotenv import load_dotenv
|
7 |
|
8 |
-
load_dotenv()
|
9 |
|
10 |
-
os.environ["HF_TOKEN"] = os.getenv("HF_TOKEN")
|
11 |
-
|
12 |
-
torch.cuda.empty_cache()
|
13 |
device = "cuda" if torch.cuda.is_available() else "cpu"
|
14 |
|
15 |
model_name = "syubraj/MedicalChat-Phi-3.5-mini-instruct"
|
@@ -31,6 +26,8 @@ def generate_response(user_query: str, system_message: str = None, max_length: i
|
|
31 |
:param max_length: Max tokens to generate.
|
32 |
:return: Generated assistant response.
|
33 |
"""
|
|
|
|
|
34 |
if not user_query.strip():
|
35 |
return "Error: User query cannot be empty."
|
36 |
|
|
|
3 |
import torch
|
4 |
import gradio as gr
|
5 |
import spaces
|
|
|
6 |
|
|
|
7 |
|
|
|
|
|
|
|
8 |
device = "cuda" if torch.cuda.is_available() else "cpu"
|
9 |
|
10 |
model_name = "syubraj/MedicalChat-Phi-3.5-mini-instruct"
|
|
|
26 |
:param max_length: Max tokens to generate.
|
27 |
:return: Generated assistant response.
|
28 |
"""
|
29 |
+
torch.cuda.empty_cache()
|
30 |
+
|
31 |
if not user_query.strip():
|
32 |
return "Error: User query cannot be empty."
|
33 |
|