Llermy

Reward is much lower when using "--play"

2

I'm training models on Mujoco environments with the PPO2 algorithm on the tf2 branch of the project. During training, reward is slowly getting higher as expected. What is not expected...

Selector with attribute matching throws exception on WebGL build

1

When it happens: At `Stylesheet stylesheet = parser.Parse(css);`, whenever the CSS string contains a selector with attribute matching and only in the WebGL build, in the editor it works fine....