DeepX-inc / machina

Control section: Deep Reinforcement Learning framework
MIT License
278 stars 45 forks source link

[pds/multi_categorical_pd.py] Fix bug of Multi categorical probabilistic distribution #215

Closed rarilurelo closed 5 years ago

rarilurelo commented 5 years ago

Last dimension of entropy in multi categorical probabilistic distribution's shape is expected to be [batch_size] or [timestep, batch_size] in loss_functional.py.