Do some preprocessing for DQN algorithm

miroesli / psscscs

Reinforcement Learning Battlesnake

MIT License

2 stars 1 forks source link

Open AhmedNSidd opened 4 years ago

AhmedNSidd commented 4 years ago

This may include:

Change input state for DQN algorithm to be similar to how Yung modified the input state so we can see snake length, snake direction etc.