Spaces:
Paused
Paused
File size: 1,614 Bytes
5be90eb 72ff248 8e90fc6 5be90eb 832ce7b 8e90fc6 2bf9d03 5be90eb 8e90fc6 5be90eb 8e90fc6 0b2a88c 5be90eb 8e90fc6 5be90eb d16c5f3 8e90fc6 5be90eb 8e90fc6 65272a9 4f5fa66 2629ae5 5be90eb |
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 63 64 65 66 |
from flask import Flask, request, jsonify
import torch
from PIL import Image
from transformers import AutoModel, AutoTokenizer, BitsAndBytesConfig
# Get API token from environment variable
api_token = os.getenv("HF_TOKEN").strip()
# Model configuration
bnb_config = BitsAndBytesConfig(
load_in_4bit=True,
bnb_4bit_quant_type="nf4",
bnb_4bit_use_double_quant=True,
bnb_4bit_compute_dtype=torch.float16,
)
# Model and tokenizer loading
model = AutoModel.from_pretrained(
"ContactDoctor/Bio-Medical-MultiModal-Llama-3-8B-V1",
quantization_config=bnb_config,
device_map="auto",
torch_dtype=torch.float16,
trust_remote_code=True,
attn_implementation="flash_attention_2",
)
tokenizer = AutoTokenizer.from_pretrained(
"ContactDoctor/Bio-Medical-MultiModal-Llama-3-8B-V1",
trust_remote_code=True
)
app = Flask(__name__)
# Model configuration and loading (same as before)
@app.route('/analyze', methods=['POST'])
def analyze():
image = request.files['image']
question = request.form['question']
# Preprocess image
image = Image.open(image).convert('RGB')
# Prepare input
msgs = [{'role': 'user', 'content': [image, question]}]
# Generate response
res = model.chat(
image=image,
msgs=msgs,
tokenizer=tokenizer,
sampling=True,
temperature=0.95,
stream=True
)
# Process response
generated_text = ""
for new_text in res:
generated_text += new_text
return jsonify({'response': generated_text})
if __name__ == '__main__':
app.run(debug=True) |