Skip to content

Latest commit

 

History

History
93 lines (67 loc) · 2.39 KB

README.md

File metadata and controls

93 lines (67 loc) · 2.39 KB
Raw

Metaflow UI Documentation

Getting started

Install dependencies with Yarn package manager.

$ yarn

Run the app in the development mode. Open http://localhost:3000 to view it in the browser.

$ yarn start

Modify Metaflow UI Service proxy and endpoints:

# Modify development proxy destination (default: http://localhost:8083)
$ METAFLOW_SERVICE_PROXY=http://localhost:8083 yarn start

# Modify both API & Websocket endpoint (default: /api via proxy)
# NOTE: Use this in case proxy is causing problems
$ REACT_APP_METAFLOW_SERVICE=http://localhost:8083 yarn start

# Modify Websocket endpoint (default: same as API with /ws suffix)
$ REACT_APP_METAFLOW_SERVICE_WS=ws://localhost:8083 yarn start

Build the app for production to the build folder.

$ yarn build

It correctly bundles React in production mode and optimizes the build for the best performance.

Docker support

Dockerfile provides support for an nginx container hosting the production build of the application.

# Build Docker image
$ docker build --tag metaflow-ui:latest .

# Run Docker container on port 3000
$ docker run -p 3000:3000 metaflow-ui:latest

# Run Docker container using custom API endpoint
$ docker run -p 3000:3000 -e METAFLOW_SERVICE=http://custom-ui-backend/api metaflow-ui:latest

Running metaflow, metaflow-service and metaflow-ui locally

# Set up metaflow service
$ git clone https://github.com/Netflix/metaflow-service.git && cd metaflow-service
# Running docker-compose.development.yml (recommended during development):
$ docker-compose -f docker-compose.development.yml up
# Run docker container using custom default API endpoint
$ docker run -p 3000:3000 -e METAFLOW_SERVICE=http://localhost:8083/ metaflow-ui:latest

# Run the metaflow code with the following custom METAFLOW_SERVICE_URL and METAFLOW_SERVICE_URL
$ METAFLOW_SERVICE_URL=http://localhost:8080/ METAFLOW_DEFAULT_METADATA=service python helloworld.py run

or

# save this in ~/.metaflowconfig/config.json
{
  "METAFLOW_DEFAULT_METADATA": "service",
  "METAFLOW_SERVICE_URL": "http://localhost:8080/"
}
# And run like a normal metaflow script
$ python helloworld.py run