Video Streamer

Live stream the training process of ML-Agents (toolkit for Reinforcement Learning with Unity Engine) using the virtual screen from Google Colab to Twitch.

Installation

!pip install mlagents-video-streamer

• Configure the video streamer.

  import videostreamer
  videostreamer.config()
  

• Activate the twitch streamer.

  xorg, ffmpeg = videostreamer.twitchStreamer('<your-twitch-secret-key>')
  

• To capture the running process, use the subprocess library.

  The below example shows how to capture the training process in the case of ML-Agents:

  import subprocess
  from random import randrange
  
  train = subprocess.run(["mlagents-learn", "config.yaml", "--run-id=train-1", "--env=3DBall_example/3DBall.x86_64", "--base-port=" + str(randrange(9000, 9999))], cwd="/content/", stdout=subprocess.PIPE)