Spaces:
Paused
Paused
Update app.py
Browse files
app.py
CHANGED
@@ -11,37 +11,38 @@ model_name = "ContactDoctor/Bio-Medical-MultiModal-Llama-3-8B-V1"
|
|
11 |
|
12 |
# Load the Hugging Face model and tokenizer with required arguments
|
13 |
tokenizer = AutoTokenizer.from_pretrained(
|
14 |
-
model_name,
|
15 |
-
|
16 |
-
trust_remote_code=True
|
17 |
)
|
18 |
model = AutoModelForCausalLM.from_pretrained(
|
19 |
-
model_name,
|
20 |
-
|
21 |
-
trust_remote_code=True
|
|
|
|
|
22 |
)
|
23 |
|
24 |
-
|
25 |
# Define the function to process user input
|
26 |
def generate_response(input_text):
|
27 |
try:
|
28 |
# Tokenize the input text
|
29 |
inputs = tokenizer(input_text, return_tensors="pt")
|
30 |
-
|
31 |
# Generate a response using the model
|
32 |
outputs = model.generate(
|
33 |
inputs["input_ids"],
|
34 |
-
max_length=256,
|
35 |
-
num_return_sequences=1,
|
36 |
-
temperature=0.7,
|
37 |
-
top_p=0.9,
|
38 |
-
top_k=50
|
39 |
)
|
40 |
-
|
41 |
# Decode and return the generated text
|
42 |
response = tokenizer.decode(outputs[0], skip_special_tokens=True)
|
43 |
return response
|
44 |
-
|
45 |
except Exception as e:
|
46 |
return f"Error: {str(e)}"
|
47 |
|
@@ -52,7 +53,7 @@ iface = gr.Interface(
|
|
52 |
outputs="text",
|
53 |
title="ContactDoctor Medical Assistant",
|
54 |
description="Provide input symptoms or queries and get AI-powered medical advice.",
|
55 |
-
enable_api=True
|
56 |
)
|
57 |
|
58 |
# Launch the Gradio app
|
|
|
11 |
|
12 |
# Load the Hugging Face model and tokenizer with required arguments
|
13 |
tokenizer = AutoTokenizer.from_pretrained(
|
14 |
+
model_name,
|
15 |
+
token=api_token, # Use `token` instead of `use_auth_token`
|
16 |
+
trust_remote_code=True
|
17 |
)
|
18 |
model = AutoModelForCausalLM.from_pretrained(
|
19 |
+
model_name,
|
20 |
+
token=api_token,
|
21 |
+
trust_remote_code=True,
|
22 |
+
device_map="auto", # Efficiently allocate resources
|
23 |
+
torch_dtype=torch.float16 # Use half precision for faster inference
|
24 |
)
|
25 |
|
|
|
26 |
# Define the function to process user input
|
27 |
def generate_response(input_text):
|
28 |
try:
|
29 |
# Tokenize the input text
|
30 |
inputs = tokenizer(input_text, return_tensors="pt")
|
31 |
+
|
32 |
# Generate a response using the model
|
33 |
outputs = model.generate(
|
34 |
inputs["input_ids"],
|
35 |
+
max_length=256,
|
36 |
+
num_return_sequences=1,
|
37 |
+
temperature=0.7,
|
38 |
+
top_p=0.9,
|
39 |
+
top_k=50
|
40 |
)
|
41 |
+
|
42 |
# Decode and return the generated text
|
43 |
response = tokenizer.decode(outputs[0], skip_special_tokens=True)
|
44 |
return response
|
45 |
+
|
46 |
except Exception as e:
|
47 |
return f"Error: {str(e)}"
|
48 |
|
|
|
53 |
outputs="text",
|
54 |
title="ContactDoctor Medical Assistant",
|
55 |
description="Provide input symptoms or queries and get AI-powered medical advice.",
|
56 |
+
enable_api=True
|
57 |
)
|
58 |
|
59 |
# Launch the Gradio app
|