Installation

Prerequisites

To run the StreamPoseML web application, you’ll need:

  • Docker and Docker Compose

  • Git (to clone the repository)

Quickstart Installation

The fastest way to get started is using the pre-built Docker images:

  1. Clone the repository:

    git clone https://github.com/mrilikecoding/StreamPoseML.git
cd StreamPoseML

  2. Start the application:

    make start

    This will:

    • Pull the necessary Docker images from Docker Hub

    • Start the containers for the API, web UI, and MLflow server

    • Launch the application in your default browser

    • Use pre-built images rather than building from source code

  3. When you’re done, stop the application:

    make stop

Additional Startup Options

For debugging purposes, you can start the application with debug output:

make start-debug

For development purposes, you can build and run from local code:

make start-dev

Local Development Installation

For developers who want to modify the web application:

  1. Clone the repository:

    git clone https://github.com/mrilikecoding/StreamPoseML.git
cd StreamPoseML

  2. Install Docker and Docker Compose:

    • Docker

    • Docker Compose is included with Docker Desktop

  3. Start local development with local code:

    make start-dev

    This will:

    • Build containers from the local source code

    • Hot-reload the API code when you make changes

    • Mount your local package code into the container

    • Provide a development environment for making changes to the code

Manual Docker Installation

If you prefer to build the Docker images manually:

# Build API image
cd stream_pose_ml && docker build -t myuser/stream_pose_ml_api:latest -f Dockerfile .

# Build web UI image
cd web_ui && docker build -t myuser/stream_pose_ml_web_ui:latest -f Dockerfile .

# Push images (if deploying to a registry)
docker push myuser/stream_pose_ml_api:latest
docker push myuser/stream_pose_ml_web_ui:latest