Refactor Docker setup: remove Ollama integration, update FastAPI service, enhance health checks, and improve README documentation

Dockerfile

@@ -1,9 +1,7 @@
-FROM ollama/ollama:0.9.2
+FROM python:3.11-slim
 
-# Install Python and dependencies with security updates
+# Install system dependencies with security updates
 RUN apt-get update && apt-get install -y \
-    python3 \
-    python3-pip \
     curl \
     && apt-get upgrade -y \
     && apt-get clean \
@@ -17,35 +15,29 @@ WORKDIR /app
 
 # Copy requirements and install Python packages
 COPY requirements.txt .
-RUN pip3 install --no-cache-dir -r requirements.txt
+RUN pip3 install --no-cache-dir --upgrade pip && \
+    pip3 install --no-cache-dir -r requirements.txt
 
 # Copy your FastAPI app
 COPY fast.py .
 COPY app.py .
-COPY start.sh .
-RUN chmod +x start.sh
 
-# Create Ollama directory and set permissions
-RUN mkdir -p /home/appuser/.ollama && \
-    chown -R appuser:appuser /home/appuser/.ollama && \
-    chmod -R 755 /home/appuser/.ollama
-
-# Change ownership to non-root user
-RUN chown -R appuser:appuser /app
+# Create logs directory and change ownership
+RUN mkdir -p /app/logs && \
+    chown -R appuser:appuser /app
 
 # Set environment variables
-ENV OLLAMA_HOST=0.0.0.0:11434
-ENV OLLAMA_ORIGINS=*
 ENV HOME=/home/appuser
-ENV OLLAMA_HOME=/home/appuser/.ollama
 
-# Expose ports
+# Expose port
 EXPOSE 7860
-EXPOSE 11434
 
 # Switch to non-root user
 USER appuser
 
-# Override the base image entrypoint and start our script
-ENTRYPOINT []
-CMD ["bash", "/app/start.sh"]
+# Health check
+HEALTHCHECK --interval=30s --timeout=10s --start-period=5s --retries=3 \
+    CMD curl -f http://localhost:7860/health || exit 1
+
+# Start FastAPI application
+CMD ["python3", "app.py"]

Dockerfile.fastapi

@@ -7,18 +7,9 @@ RUN apt-get update && apt-get install -y \
     && apt-get clean \
     && rm -rf /var/lib/apt/lists/* /tmp/* /var/tmp/*
 
-# Install Ollama
-RUN curl -fsSL https://ollama.com/install.sh | sh
-
 # Create non-root user for security
 RUN groupadd -r appuser && useradd -r -g appuser -m appuser
 
-# Create Ollama directories with proper permissions
-RUN mkdir -p /home/appuser/.ollama && \
-    mkdir -p /home/appuser/.ollama/models && \
-    chown -R appuser:appuser /home/appuser/.ollama && \
-    chmod -R 755 /home/appuser/.ollama
-
 # Set working directory
 WORKDIR /app
 
@@ -27,17 +18,14 @@ COPY requirements.txt .
 RUN pip3 install --no-cache-dir --upgrade pip && \
     pip3 install --no-cache-dir -r requirements.txt
 
-# Copy FastAPI application and startup script
+# Copy FastAPI application
 COPY fast.py .
-COPY fastapi_startup_script.sh .
-RUN chmod +x fastapi_startup_script.sh
 
 # Create logs directory and change ownership
 RUN mkdir -p /app/logs && \
     chown -R appuser:appuser /app
 
-# Set environment variables for Ollama
-ENV OLLAMA_HOME=/home/appuser/.ollama
+# Set environment variables
 ENV HOME=/home/appuser
 
 # Switch to non-root user
@@ -50,5 +38,5 @@ EXPOSE 7860
 HEALTHCHECK --interval=30s --timeout=10s --start-period=5s --retries=3 \
     CMD curl -f http://localhost:7860/health || exit 1
 
-# Start Ollama and FastAPI
-CMD ["./fastapi_startup_script.sh"]
+# Start FastAPI application
+CMD ["python3", "-m", "uvicorn", "fast:app", "--host", "0.0.0.0", "--port", "7860"]

Makefile

@@ -14,7 +14,6 @@ help:
 	@echo "  status    - Show service status"
 	@echo "  health    - Check application health"
 	@echo "  shell-api - Open shell in FastAPI container"
-	@echo "  shell-ollama - Open shell in Ollama container"
 
 # Build all services
 build:
@@ -48,8 +47,6 @@ status:
 
 # Check application health
 health:
-	@echo "Checking Ollama health..."
-	@curl -s http://localhost:11434/api/tags > /dev/null && echo "✅ Ollama: Healthy" || echo "❌ Ollama: Unhealthy"
 	@echo "Checking FastAPI health..."
 	@curl -s http://localhost:7860/health > /dev/null && echo "✅ FastAPI: Healthy" || echo "❌ FastAPI: Unhealthy"
 
@@ -57,16 +54,9 @@ health:
 shell-api:
 	docker-compose exec fastapi bash
 
-# Open shell in Ollama container
-shell-ollama:
-	docker-compose exec ollama bash
-
 # Development commands
 dev-build:
 	docker-compose build --no-cache
 
 dev-logs-api:
 	docker-compose logs -f fastapi
-
-dev-logs-ollama:
-	docker-compose logs -f ollama

README-docker-compose.md

@@ -1,26 +1,25 @@
 # Fashion Analyzer - Docker Compose Setup
 
-This project provides a secure, containerized fashion analysis application using Ollama and FastAPI with Docker Compose.
+This project provides a secure, containerized fashion analysis application using Hugging Face Transformers and FastAPI with Docker Compose.
 
 ## 🏗️ Architecture
 
-The application consists of three services:
+The application consists of a single service:
 
-1. **Ollama Service**: Runs the Ollama server with LLaVA model for vision analysis
-2. **FastAPI Service**: Provides the web API and user interface
-3. **Model Loader**: One-time service to download the required LLaVA model
+1. **FastAPI Service**: Provides the web API and user interface with integrated Hugging Face transformers models for fashion analysis
 
 ## 🔒 Security Features
 
-- ✅ **Pinned Ollama version** (0.9.2) - No critical vulnerabilities
+- ✅ **Pinned Python version** (3.11-slim) - Secure base image
 - ✅ **Non-root user execution** - Enhanced container security
 - ✅ **Security updates** - Latest package updates applied
 - ✅ **Health checks** - Service monitoring and restart policies
-- ✅ **Network isolation** - Services communicate via internal network
+- ✅ **Minimal dependencies** - Reduced attack surface
 
 ## 🚀 Quick Start
 
 ### Prerequisites
+
 - Docker Engine 20.10+
 - Docker Compose 2.0+
 
@@ -42,7 +41,6 @@ docker-compose ps
 - **Web Interface**: http://localhost:7860
 - **API Documentation**: http://localhost:7860/docs
 - **Health Check**: http://localhost:7860/health
-- **Ollama API**: http://localhost:11434
 
 ### 3. Stop the Application
 
@@ -50,7 +48,7 @@ docker-compose ps
 # Stop all services
 docker-compose down
 
-# Stop and remove volumes (removes downloaded models)
+# Stop and remove volumes (removes logs)
 docker-compose down -v
 ```
 
@@ -72,11 +70,9 @@ AI/
 ### Environment Variables (.env)
 
 ```env
-OLLAMA_HOST=0.0.0.0:11434
-OLLAMA_ORIGINS=*
-OLLAMA_BASE_URL=http://ollama:11434
-OLLAMA_PORT=11434
 FASTAPI_PORT=7860
+ENVIRONMENT=development
+LOG_LEVEL=info
 ```
 
 ### Custom Configuration
@@ -106,7 +102,6 @@ docker-compose logs -f
 
 # Specific service
 docker-compose logs -f fastapi
-docker-compose logs -f ollama
 ```
 
 ### Debugging
@@ -114,7 +109,6 @@ docker-compose logs -f ollama
 ```bash
 # Execute commands in running containers
 docker-compose exec fastapi bash
-docker-compose exec ollama bash
 
 # Check service health
 docker-compose exec fastapi curl http://localhost:7860/health
@@ -124,22 +118,17 @@ docker-compose exec fastapi curl http://localhost:7860/health
 
 ### Health Checks
 
-All services include health checks:
-- **Ollama**: Checks API availability
-- **FastAPI**: Checks application health and Ollama connectivity
-
-### Service Dependencies
+The FastAPI service includes health checks:
 
-- FastAPI waits for Ollama to be healthy before starting
-- Model loader runs after Ollama is ready
+- **FastAPI**: Checks application health and model availability
 
 ## 🛠️ Troubleshooting
 
 ### Common Issues
 
 1. **Port conflicts**: Change ports in `.env` file
-2. **Model download fails**: Check internet connection and Ollama logs
-3. **FastAPI can't connect to Ollama**: Verify network configuration
+2. **Model loading fails**: Check internet connection and available memory
+3. **Application startup slow**: Transformers models need time to download and load
 
 ### Reset Everything
 

docker-compose.prod.yml

@@ -1,31 +1,15 @@
-version: '3.8'
-
 # Production overrides for docker-compose.yml
 # Usage: docker-compose -f docker-compose.yml -f docker-compose.prod.yml up -d
 
 services:
-  ollama:
-    restart: always
-    deploy:
-      resources:
-        limits:
-          memory: 8G
-        reservations:
-          memory: 4G
-    logging:
-      driver: "json-file"
-      options:
-        max-size: "10m"
-        max-file: "3"
-
   fastapi:
     restart: always
     deploy:
       resources:
         limits:
-          memory: 1G
+          memory: 4G # Increased for transformers models
         reservations:
-          memory: 512M
+          memory: 2G
     logging:
       driver: "json-file"
       options:

docker-compose.yml

@@ -1,26 +1,4 @@
-version: "3.8"
-
 services:
-  ollama:
-    image: ollama/ollama:0.9.2
-    container_name: ollama-server
-    restart: unless-stopped
-    ports:
-      - "11434:11434"
-    volumes:
-      - ollama_data:/root/.ollama
-    environment:
-      - OLLAMA_HOST=0.0.0.0:11434
-      - OLLAMA_ORIGINS=*
-    healthcheck:
-      test: ["CMD", "curl", "-f", "http://localhost:11434/api/tags"]
-      interval: 30s
-      timeout: 10s
-      retries: 3
-      start_period: 30s
-    networks:
-      - fashion-analyzer
-
   fastapi:
     build:
       context: .
@@ -29,38 +7,11 @@ services:
     restart: unless-stopped
     ports:
       - "7860:7860"
-    environment:
-      - OLLAMA_BASE_URL=http://ollama:11434
-    depends_on:
-      - ollama
     networks:
       - fashion-analyzer
     volumes:
       - ./logs:/app/logs
 
-  model-loader:
-    image: ollama/ollama:0.9.2
-    container_name: model-loader
-    restart: "no"
-    environment:
-      - OLLAMA_HOST=http://ollama:11434
-    depends_on:
-      - ollama
-    networks:
-      - fashion-analyzer
-    command: >
-      sh -c "
-        echo 'Waiting for Ollama server to be ready...' &&
-        sleep 10 &&
-        echo 'Pulling LLaVA model for vision analysis...' &&
-        ollama pull llava:7b &&
-        echo 'Model pulled successfully!'
-      "
-
-volumes:
-  ollama_data:
-    driver: local
-
 networks:
   fashion-analyzer:
     driver: bridge

fastapi_startup_script.sh

@@ -3,27 +3,9 @@
 # Debug: Check current user and permissions
 echo "Current user: $(whoami)"
 echo "Home directory: $HOME"
-echo "Ollama home: $OLLAMA_HOME"
-echo "Checking .ollama directory permissions:"
-ls -la /home/appuser/.ollama/ || echo "Directory doesn't exist"
 
-# Start Ollama server in background
-echo "Starting Ollama server..."
-ollama serve &
-
-# Wait for Ollama to start
-echo "Waiting for Ollama server to start..."
-while ! curl -s http://localhost:11434 > /dev/null; do
-  sleep 1
-done
-
-echo "Ollama server started!"
-
-# Pull the LLaVA model for vision analysis
-echo "Pulling LLaVA model for vision analysis..."
-ollama pull llava:7b
-
-echo "Model pulled successfully!"
+# Note: This script is deprecated as the application now uses transformers instead of Ollama
+echo "Starting FastAPI server with transformers-based fashion analysis..."
 
 # Start FastAPI on port 7860 (HF Spaces requirement)
 echo "Starting FastAPI server on port 7860..."

start.sh

@@ -6,66 +6,10 @@ echo "Current user: $(whoami)"
 echo "User ID: $(id)"
 echo "Current directory: $(pwd)"
 echo "Home directory: $HOME"
-echo "Ollama home: $OLLAMA_HOME"
 echo "PATH: $PATH"
-echo "Which ollama: $(which ollama)"
-echo "Ollama version: $(ollama --version 2>&1 || echo 'Failed to get version')"
-
-# Check Ollama directory permissions
-echo "Checking Ollama directory:"
-ls -la $HOME/.ollama/ 2>/dev/null || echo "Ollama directory doesn't exist"
-echo "Checking if we can write to Ollama directory:"
-touch $HOME/.ollama/test_write 2>/dev/null && rm $HOME/.ollama/test_write && echo "Write test: SUCCESS" || echo "Write test: FAILED"
 echo "========================="
 
-# Ensure Ollama directory exists and has correct permissions
-echo "Setting up Ollama directory..."
-mkdir -p $HOME/.ollama
-chmod 755 $HOME/.ollama
-echo "Ollama directory setup complete."
-
-# Start Ollama server in background
-echo "Starting Ollama server..."
-
-# Try different ways to start Ollama in case of PATH issues
-if command -v ollama >/dev/null 2>&1; then
-    echo "Using ollama from PATH"
-    ollama serve &
-elif [ -f "/usr/local/bin/ollama" ]; then
-    echo "Using ollama from /usr/local/bin/"
-    /usr/local/bin/ollama serve &
-elif [ -f "/usr/bin/ollama" ]; then
-    echo "Using ollama from /usr/bin/"
-    /usr/bin/ollama serve &
-else
-    echo "ERROR: Could not find ollama binary!"
-    exit 1
-fi
-
-# Wait for Ollama to start
-echo "Waiting for Ollama server to start..."
-while ! curl -s http://localhost:11434 > /dev/null; do
-  sleep 1
-done
-
-echo "Ollama server started!"
-
-# Pull the LLaVA model for vision analysis
-echo "Pulling LLaVA model for vision analysis..."
-
-# Use the same approach for pulling models
-if command -v ollama >/dev/null 2>&1; then
-    ollama pull llava:7b
-elif [ -f "/usr/local/bin/ollama" ]; then
-    /usr/local/bin/ollama pull llava:7b
-elif [ -f "/usr/bin/ollama" ]; then
-    /usr/bin/ollama pull llava:7b
-else
-    echo "ERROR: Could not find ollama binary for pulling model!"
-    exit 1
-fi
-
-echo "Model pulled successfully!"
+echo "Starting Fashion Analyzer with transformers-based AI models..."
 
 # Start FastAPI on port 7860 (HF Spaces requirement)
 echo "Starting FastAPI server on port 7860..."