google-research / batch-ppo

Efficient Batched Reinforcement Learning in TensorFlow

google-research/batch-ppo Issues

Performance issue in agents/parts/iterate_sequences.py
Closed 3 years ago2
Performance issues in the program
Closed 3 years ago1
With GPU enabled, tensorflow freezes unless I force "Discounted Monte-Carlo returns." to CPU
Updated 6 years ago3
Generalized advantage estimation
Closed 6 years ago4
How to draw the score curve of PPO?
Closed 6 years ago6
Training threads don't start on Windows
Updated 6 years ago19
OpenAI Retro environment support/example
Closed 6 years ago3
Minimal Example For TPU?
Closed 6 years ago3
setup.py calls this module agents and refers to tensorflow/agents
Closed 6 years ago1
Maximum number of clients reachedSegmentation fault (core dumped)
Closed 6 years ago4
Errors about the regularizer term
Closed 6 years ago2
Computation time needed for HalfCheetah
Closed 7 years ago2
Enabling GPU compromises learning performance
Closed 7 years ago9
Distributed training with Kubernetes
Updated 6 years ago28
No initial states in update_steps
Closed 7 years ago1
Potential integration of MonitoredTrainingSession and revision of tools/loop.py
Closed 7 years ago
Running out of GPU memory
Closed 6 years ago5
Support for discrete action spaces
Closed 6 years ago3
Typo in Paper
Closed 7 years ago1
GPU doesn't seem to work
Updated 6 years ago9
No module named 'mujoco_py.mjlib'
Closed 7 years ago2
Different frames error when adding another RNNCell
Closed 7 years ago2
Modification of network: How to handle multiple inputs
Closed 7 years ago7