TensorFlow: Symbolic entropy for DiagonalGaussian

rll / rllab

rllab is a framework for developing and evaluating reinforcement learning algorithms, fully compatible with OpenAI Gym.

Other

2.91k stars 800 forks source link

Open ryanjulian opened 6 years ago

ryanjulian commented 6 years ago

Adds a symbolic entropy operation to the DiagonalGaussian implementation in the TensorFlow tree. This is equivalent to https://github.com/rll/rllab/blob/master/rllab/distributions/diagonal_gaussian.py#L89