Spaces:
Paused
Paused
File size: 1,750 Bytes
882bd69 832ce7b 882bd69 fa9231d 5f6d422 832ce7b fa9231d 882bd69 832ce7b fa9231d 9998c92 fa9231d 9998c92 fa9231d 882bd69 9998c92 882bd69 9998c92 882bd69 9998c92 882bd69 9998c92 882bd69 9998c92 882bd69 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 |
import gradio as gr
from transformers import AutoTokenizer, AutoModelForCausalLM
import torch
import os
# Retrieve the token from environment variables
api_token = os.getenv("HF_TOKEN").strip()
# Model name
model_name = "ContactDoctor/Bio-Medical-MultiModal-Llama-3-8B-V1"
# Load the Hugging Face model and tokenizer with required arguments
tokenizer = AutoTokenizer.from_pretrained(
model_name,
token=api_token, # Use `token` instead of `use_auth_token`
trust_remote_code=True
)
model = AutoModelForCausalLM.from_pretrained(
model_name,
token=api_token,
trust_remote_code=True,
device_map="auto", # Efficiently allocate resources
torch_dtype=torch.float16 # Use half precision for faster inference
)
# Define the function to process user input
def generate_response(input_text):
try:
# Tokenize the input text
inputs = tokenizer(input_text, return_tensors="pt")
# Generate a response using the model
outputs = model.generate(
inputs["input_ids"],
max_length=256,
num_return_sequences=1,
temperature=0.7,
top_p=0.9,
top_k=50
)
# Decode and return the generated text
response = tokenizer.decode(outputs[0], skip_special_tokens=True)
return response
except Exception as e:
return f"Error: {str(e)}"
# Create a Gradio interface with API enabled
iface = gr.Interface(
fn=generate_response,
inputs="text",
outputs="text",
title="ContactDoctor Medical Assistant",
description="Provide input symptoms or queries and get AI-powered medical advice.",
enable_api=True
)
# Launch the Gradio app
if __name__ == "__main__":
iface.launch()
|