 Maxime Chevalier-Boisvert
|
c99822121e
Added reward penalty based on number of time steps taken
|
vor 6 Jahren |
|
Maxime Chevalier-Boisvert
|
146fd10741
Made reward_range the same for all environments, rewards are in [0, 1]
|
vor 6 Jahren |
|
Maxime Chevalier-Boisvert
|
852476db7c
Finished renaming MiniGrid methods for PEP8 conformance
|
vor 6 Jahren |
|
Maxime Chevalier-Boisvert
|
bfd0f76513
Faster visibility algorithm. Method renamings.
|
vor 6 Jahren |
|
Maxime Chevalier-Boisvert
|
ec9e19efe7
Renamed fields to match PEP8 convention
|
vor 6 Jahren |
|
Maxime Chevalier-Boisvert
|
340c03a446
Cleaned up and simplified _genGrid functions
|
vor 6 Jahren |
|
Maxime Chevalier-Boisvert
|
2cdc42ac43
Modified reward range for fetch environment
|
vor 6 Jahren |
|
Maxime Chevalier-Boisvert
|
25fe4664fa
Modified environments so they all produce observations in a dict
|
vor 6 Jahren |
|
Maxime Chevalier-Boisvert
|
114caa944a
Fixes based on changes in OpenAI Gym 0.9.6
|
vor 6 Jahren |
|
Maxime Chevalier-Boisvert
|
c46ade2f4f
Fixed bug in fetch environment
|
vor 6 Jahren |
|
Maxime Chevalier-Boisvert
|
3d0c94f876
Removed "advice" from observations. Randomized GoToDoor room size.
|
vor 6 Jahren |
|
Maxime Chevalier-Boisvert
|
32535cc145
Changed reward range for fetch environment
|
vor 6 Jahren |
|
Maxime Chevalier-Boisvert
|
e02777d81b
Added function to encode observations to FetchEnv
|
vor 6 Jahren |
|
Maxime Chevalier-Boisvert
|
16698be044
Added 5x5 config for the fetch environment
|
vor 6 Jahren |
|
Maxime Chevalier-Boisvert
|
5daf219a68
Added tests. Moved envs into own source files.
|
vor 6 Jahren |
