Miniracer_traning

This is about CNN; convolutional autoencoder and RL agent

1. Train a CNN to predict a clear road ahead
2. Train a convolutional autoencoder
3. Create a RL agent for Minirace (level 1)
4. Create a RL agent for Minirace (level 2)

In this project Reinforcement Learning part, I used Kyoke as my library. For the details: https://ishikota.github.io/kyoka/

+2 for the car (the dark 2 × 2 square), 0 for drivable track segments (1 × 6 pixels, in white), and +1 for non-drivable terrain (here in grey)

Running sprites.py will create datasets of screenshots for CNN and autoencoder task