gryf/coach
1
0
Fork 0
mirror of https://github.com/gryf/coach.git synced 2026-04-03 18:53:31 +02:00
Code Issues Projects Releases Wiki Activity

Batch RL Tutorial (#372)

Browse Source
This commit is contained in:
Gal Leibovich
2019-07-14 18:43:48 +03:00
committed by GitHub
parent b82414138d
commit 19ad2d60a7
40 changed files with 1155 additions and 182 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
README.md
View File

@@ -169,8 +169,13 @@ More usage examples can be found [here](https://github.com/NervanaSystems/coach/
### Distributed Multi-Node Coach
As of release 0.11 Coach supports horizontal scaling for training RL agents on multiple nodes. In release 0.11 this was tested on the ClippedPPO and DQN agents.
For usage instructions please refer to the documentation [here](https://nervanasystems.github.io/coach/dist_usage.html)
As of release 0.11.0, Coach supports horizontal scaling for training RL agents on multiple nodes. In release 0.11.0 this was tested on the ClippedPPO and DQN agents.
For usage instructions please refer to the documentation [here](https://nervanasystems.github.io/coach/dist_usage.html).
### Batch Reinforcement Learning
Training and evaluating an agent from a dataset of experience, where no simulator is available, is supported in Coach.
There are [example](https://github.com/NervanaSystems/coach/blob/master/rl_coach/presets/CartPole_DDQN_BatchRL.py) [presets](https://github.com/NervanaSystems/coach/blob/master/rl_coach/presets/Acrobot_DDQN_BCQ_BatchRL.py) and a [tutorial](https://github.com/NervanaSystems/coach/blob/master/tutorials/4.%20Batch%20Reinforcement%20Learning.ipynb).
### Running Coach Dashboard (Visualization)
Training an agent to solve an environment can be tricky, at times.