gryf/coach
1
0
Fork 0
mirror of https://github.com/gryf/coach.git synced 2025-12-17 11:10:20 +01:00
Code Issues Projects Releases Wiki Activity

Update README.md

Browse Source
This commit is contained in:
Gal Leibovich
2018-08-13 17:19:47 +03:00
committed by GitHub
parent 19ca5c24b1
commit 7a76d63da4
1 changed files with 5 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
README.md
View File

@@ -111,20 +111,20 @@ To list all the available presets use the `-l` flag.
To run a preset, use: To run a preset, use:
```bash ```bash
python3 coach.py -r -p <preset_name> python3 rl_coach/coach.py -r -p <preset_name>
``` ```
For example: For example:
* CartPole environment using Policy Gradients (PG): * CartPole environment using Policy Gradients (PG):
```bash ```bash
python3 coach.py -r -p CartPole_PG python3 rl_coach/coach.py -r -p CartPole_PG
``` ```
* Basic level of Doom using Dueling network and Double DQN (DDQN) algorithm: * Basic level of Doom using Dueling network and Double DQN (DDQN) algorithm:
```bash ```bash
python3 coach.py -r -p Doom_Basic_Dueling_DDQN python3 rl_coach/coach.py -r -p Doom_Basic_Dueling_DDQN
``` ```
Some presets apply to a group of environment levels, like the entire Atari or Mujoco suites for example. Some presets apply to a group of environment levels, like the entire Atari or Mujoco suites for example.
@@ -136,7 +136,7 @@ For example:
* Pong using the Nerual Episodic Control (NEC) algorithm: * Pong using the Nerual Episodic Control (NEC) algorithm:
```bash ```bash
python3 coach.py -r -p Atari_NEC -lvl pong python3 rl_coach/coach.py -r -p Atari_NEC -lvl pong
``` ```
There are several types of agents that can benefit from running them in a distrbitued fashion with multiple workers in parallel. Each worker interacts with its own copy of the environment but updates a shared network, which improves the data collection speed and the stability of the learning process. There are several types of agents that can benefit from running them in a distrbitued fashion with multiple workers in parallel. Each worker interacts with its own copy of the environment but updates a shared network, which improves the data collection speed and the stability of the learning process.
@@ -146,7 +146,7 @@ For example:
* Breakout using Asynchronous Advantage Actor-Critic (A3C) with 8 workers: * Breakout using Asynchronous Advantage Actor-Critic (A3C) with 8 workers:
```bash ```bash
python3 coach.py -r -p Atari_A3C -lvl breakout -n 8 python3 rl_coach/coach.py -r -p Atari_A3C -lvl breakout -n 8
``` ```