Spaces:
Sleeping
Sleeping
File size: 1,409 Bytes
1ea1964 |
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 |
from fastapi import FastAPI, File, UploadFile
import torch
from transformers import CLIPProcessor, CLIPModel
from dotenv import load_dotenv
import logging
import os
from PIL import Image
load_dotenv()
logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)
app = FastAPI(title="Image Embedding API",
description="Returns CLIP image embeddings via GET")
HF_TOKEN = os.getenv('hf_token')
logger.info("Loading CLIP processor and model...")
try:
processor = CLIPProcessor.from_pretrained(
"openai/clip-vit-large-patch14", use_auth_token=HF_TOKEN)
clip_model = CLIPModel.from_pretrained(
"openai/clip-vit-large-patch14", use_auth_token=HF_TOKEN)
clip_model.eval()
logger.info("CLIP model loaded successfully")
except Exception as e:
logger.error(f"Failed to load CLIP model: {e}")
raise
@app.get("/")
async def root():
logger.info("Root endpoint accessed")
return {"message": "Welcome to the Image Embedding API."}
@app.post("/clip/process")
async def process_image(file: UploadFile = File(...)):
logger.info("Processing image")
image = Image.open(file.file).convert("RGB")
inputs = processor(images=image, return_tensors="pt")
with torch.no_grad():
embeddings = clip_model.get_image_features(**inputs)
return {"embedding": embeddings.tolist()}
|