pommerlearn - Initial Release with Models

Attached are * 1M_simple_0.zr.zip: Exemplary training data set with FFA games played by four SimpleUnbiasedAgent agents * models_sl_dummy_onnx.zip: Dummy models and models trained on that data set with SL (onnx) * models_rl14.zip: Models trained with RL and single-player search (from dummy and from sl) for 15 iterations (onnx)

- Python
Published by jw3il almost 3 years ago