KoalaTrapLord
52cbefaccf
Adds placeholder structure for all Phase 0-6 modules. See issues #167-#174 for full implementation details.
5 lines
156 B
Python
5 lines
156 B
Python
# Phase 6 - training_data_builder.py
|
|
# Converts decision_log + game_result -> (state, action, reward) tuples
|
|
# Feeds existing PPO pipeline
|
|
# See issue #173
|