PPO Config

Browser-based PPO implementation.

Uses TensorFlow.js for training.

CartPole-v1

IDLE
Global Step: 0
Updates: 0
Goal: Balance the pole. +1 Reward/Step.
Waiting for data...
Avg Reward (Last 10)
0.0
Best Reward
-
Ready to start training...