GIA RunwayML HTTP

GIA RunwayML HTTP (RunwayHTTP) is a utility package that provides quick and easy methods to post and request data from the RunwayML desktop app.

Demos

(Coming soon)

Installation

npm i -S gia-runwayml-http

Usage (Node.js + npm)

Use the following syntax to add RunwayHTTP to the project:

const GIARunwayHTTP = require('gia-runway-http');

// OR alternative ES6 import syntax
import GIARunwayHTTP from 'gia-runwayml-http';

In your selected RunwayML workspace, navigate to Network > HTTP to check the HTTP server address port (Server Address default port is 8000). Input Specification also shows the correct data format expected by RunwayML.

Send data to RunwayML:

// Query server with port and data.
const captionData = { 'caption': 'Machine learning for creators' };

GIARunwayHTTP.query("192.168.2.44",8000, captionData);

The query method returns a response from RunwayML:

// Query server and handle response using a Promise.
const captionData = { 'caption': 'Machine learning for creators' };

GIARunwayHTTP.query("192.168.2.44",8000, captionData)
    .then(res => console.log(res));

RunwayHTTP also includes getData, getInfo, and getError methods that correspond to RunwayML's routes:

// GET from /data route
GIARunwayHTTP.getData("localhost",8001).then(res => console.log('data:', res));

// GET from /info route
GIARunwayHTTP.getInfo("192.168.2.44",8000).then(res => console.log('info:', res));

// GET from /error route
GIARunwayHTTP.getError("192.168.2.44",8000).then(res => console.log('error:', res));

Release Notes

1.0.0

• Initial Release

Credits

This package has been implemented according to RunwayML's documentation.

Author: JP Yepez

Forked: GuillaumeAI