Installation

DeepChem Server can be installed and run using Docker (recommended) or manually from source. Docker provides the simplest setup with all dependencies pre-configured.

Docker Installation (Recommended)

Docker provides containerized deployment with all dependencies managed automatically.

Prerequisites

• Docker installed on your system

• Docker Compose (included with Docker Desktop)

Steps

1. Clone the Repository

   git clone <repository-url>
   cd deepchem-server
   

2. Start the Server

   Use the provided helper script for easy setup:

   ./docker.sh
   

   Available Options:

   ./docker.sh
   

   Examples:

   # Use default CPU setup
   ./docker.sh
   
   # Use GPU setup
   ./docker.sh -f Dockerfile.gpu
   

3. Verify Installation

   Check that the server is running:

   curl http://localhost:8000/healthcheck
   

   Expected response: {"status": "ok"}

   Access the interactiveAPI documentation at: http://localhost:8000/docs

Available Dockerfiles

• Dockerfile: CPU-based setup using micromamba base image

• Dockerfile.gpu: GPU-accelerated setup using NVIDIA PyTorch base image

Dockerfile Features:

• Uses micromamba for fast package management

• Includes all DeepChem dependencies

• Optimized for CPU-based molecular machine learning

• Health check endpoint configured

Dockerfile.gpu Features:

• NVIDIA PyTorch base image with CUDA support

• GPU-accelerated DeepChem operations

• Compatible with NVIDIA Docker runtime

• Optimized for large-scale molecular modeling

Manual Installation

For development or custom deployments, install DeepChem Server manually using micromamba.

Prerequisites

• micromamba package manager (lightweight conda replacement)

• Python 3.11 (specified in environment file)

• Git for cloning the repository

Steps

1. Clone the Repository

   git clone <repository-url>
   cd deepchem-server
   

2. Create Environment from YAML

   Use the provided environment file to create a consistent environment:

   micromamba env create -f deepchem_server/environments/core_environment.yml
   

3. Activate Environment

   micromamba activate deepchem-server-env
   

4. Start the Server

   For development with auto-reload:

   cd deepchem_server
   uvicorn main:app --host 0.0.0.0 --port 8000 --reload
   

   Or use the development script:

   ./start-dev-server.sh
   

Configuration

Environment Management

• Environment File: deepchem_server/environments/core_environment.yml

• Python Version: 3.11

• Package Manager: micromamba for dependency management

Environment Variables

• DATADIR: Data directory path (default: /opt/deepchem_server_app/data in Docker)

• PYTHONPATH: Python path for module imports

Port Configuration

Default port is 8000. To change:

# Docker
docker run -p 8080:8000 deepchem-server

# Manual
uvicorn main:app --port 8080

Verification

After installation, verify the setup:

1. Health Check

   curl http://localhost:8000/healthcheck
   

2. API Documentation

   Visit http://localhost:8000/docs for interactive API documentation

3. Test Upload

   cd pyds/tests
   python test_upload_featurize.py
   

Getting Help

If you encounter issues:

1. Check server logs for error messages

2. Verify micromamba environment is correctly created and activated

3. Ensure system meets minimum requirements

4. Verify Python version is 3.11 as specified in environment file

5. Consult the FastAPI documentation for deployment questions

6. Check the DeepChem documentation for primitive operations