deploy

Dockerfile

@@ -1,7 +1,23 @@
+# Use official Python 3.9 image
 FROM python:3.10-slim
+
+# Set working directory
 WORKDIR /app
+
+# Copy requirements file
 COPY requirements.txt .
+
+# Install dependencies
 RUN pip install --no-cache-dir -r requirements.txt
-COPY . . 
+
+# Create cache directory with proper permissions
+RUN mkdir -p /app/cache && chmod -R 777 /app/cache
+
+# Copy application code
+COPY . .
+
+# Expose port 8000
 EXPOSE 7860
+
+# Start FastAPI with uvicorn
 CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "7860"]

services/embedding.py

@@ -3,8 +3,15 @@ from PIL import Image
 from fastapi import UploadFile
 from typing import List, Optional
 import torch
+import os
 
-model = SentenceTransformer("clip-ViT-B-32")
+# model = SentenceTransformer("clip-ViT-B-32")
+# Set a writable cache directory
+os.environ["TRANSFORMERS_CACHE"] = "/app/cache"
+os.environ["HF_HOME"] = "/app/cache"
+
+# Load model with custom cache directory
+model = SentenceTransformer("clip-ViT-B-32", cache_dir="/app/cache")
 
 def get_text_embedding(text: str) -> Optional[List[float]]:
     try: