coach
coach copied to clipboard
Reinforcement Learning Coach by Intel AI Lab enables easy experimentation with state of the art Reinforcement Learning algorithms
I've been using rllib for a while and quite like the [Trainable](https://ray.readthedocs.io/en/latest/tune-advanced-tutorial.html#a-native-example-of-trainable) and Tune (hyperaram search) abstractions. Does coach contain something similar? I really like coach's documentation. Can someone provide...
Coach has only been tested on Ubuntu 16.04 LTS and with Python 3.5 while mujoco-py supports Linux with Python 3.6+.
Greetings, I downloaded version 1.0.0 and tried running the presets CARLA_DDPG and CARLA_DOUBLE_DDQN in multi-thread mode but I'm getting TCP connection issues between client and server. Here are some details:...
That's kinda weird, any idea on why is that? (I did not modify anything in the codebase)
I realize my problem set might be too complicated for the agent to learn. Is there a way to assist it to learn like maybe inputting a correct solution?
Good day, I am experiencing add issue while testing a simple custom environment with default DQN agent. I have state space `PlanarMapsObservationSpace(shape=np.array([50, 50, 1]), low=0, high=1)` and discrete action space...
Hi all, During my tries with `rl_coach,` I have used the `checkpoint_save_secs` parameter to save intermediate checkpoints of my training. I have two question regarding this parameter, that I may...
Hi, In my experiments, I wanted to use multiple observation inputs. In the preset, I defined it to be: ``` agent_params = ClippedPPOAgentParameters() agent_params.network_wrappers['main'].input_embedders_parameters = { 'front_camera': InputEmbedderParameters(), 'lidar': InputEmbedderParameters()}...
Hedges against CircleCi automatic re-run behaviour causing failure owing to the namespace being created already existing