 Lucas Willems
|
c125ca7998
Minimum reward when success is now 0.1 (#9)
|
il y a 7 ans |
 Maxime Chevalier-Boisvert
|
290ab259e4
Modified RedBlueDoor env to enforce door opening sequence
|
il y a 7 ans |
|
Maxime Chevalier-Boisvert
|
c99822121e
Added reward penalty based on number of time steps taken
|
il y a 7 ans |
|
Maxime Chevalier-Boisvert
|
636226f90e
Renamed wait action to done
|
il y a 7 ans |
|
Maxime Chevalier-Boisvert
|
0ad70c61dd
Added _rand_subset method to MiniGridEnv
|
il y a 7 ans |
|
Maxime Chevalier-Boisvert
|
40f29632c2
Added ability to render observations at any resolution
|
il y a 7 ans |
|
Maxime Chevalier-Boisvert
|
041225e96b
Added position randomization to RedBlueDoors env. Updated README.
|
il y a 7 ans |
|
Maxime Chevalier-Boisvert
|
216fe87800
Merge pull request #8 from lcswillems/master
|
il y a 7 ans |
 Lucas Willems
|
82d080fa8b
Merge remote-tracking branch 'upstream/master'
|
il y a 7 ans |
|
Lucas Willems
|
ec123f87cf
Add a RedBlueDoors environment
|
il y a 7 ans |
|
Maxime Chevalier-Boisvert
|
2b243b906f
Update README.md
|
il y a 7 ans |
|
Maxime Chevalier-Boisvert
|
342b3c96f1
Fixed bug in place_agent
|
il y a 7 ans |
|
Maxime Chevalier-Boisvert
|
5d88cb4376
Doors can now be closed back after they are opened
|
il y a 7 ans |
|
Maxime Chevalier-Boisvert
|
4f4992265b
Fixed issue wrt agent view size pointed out by Anirudh
|
il y a 7 ans |
|
Maxime Chevalier-Boisvert
|
2190ce0b59
Full map correctly highlights cells visible to the agent
|
il y a 7 ans |
|
Maxime Chevalier-Boisvert
|
076e074b0a
Changed timestep limits based on feedback from Rosemary Ke
|
il y a 7 ans |
|
Maxime Chevalier-Boisvert
|
6b513362e6
Added gen_obs_grid method which outputs visibility mask
|
il y a 7 ans |
|
Maxime Chevalier-Boisvert
|
a6ece826ce
Removed old visibility code
|
il y a 7 ans |
|
Maxime Chevalier-Boisvert
|
76b43b7534
Added DIR_TO_VEC array. Agent position is now a numpy array.
|
il y a 7 ans |
|
Maxime Chevalier-Boisvert
|
d5d117bb78
Added goal_pos variable to MultiRoomEnv
|
il y a 7 ans |
|
Maxime Chevalier-Boisvert
|
146fd10741
Made reward_range the same for all environments, rewards are in [0, 1]
|
il y a 7 ans |
|
Maxime Chevalier-Boisvert
|
fe469bb1cc
Added MiniGrid._rand_color method
|
il y a 7 ans |
|
Maxime Chevalier-Boisvert
|
30c97ffc57
Added _rand_bool() method to MiniGrid
|
il y a 7 ans |
|
Maxime Chevalier-Boisvert
|
d562fb8f82
Fixup. Default random seed should be fixed, deterministic.
|
il y a 7 ans |
|
Maxime Chevalier-Boisvert
|
9b585b9b51
Added seed argument to MiniGrid constructor, removed RoomGrid
|
il y a 7 ans |
|
Maxime Chevalier-Boisvert
|
da1e2c5c5e
Increased max_steps on Empty & DoorKey envs
|
il y a 7 ans |
|
Maxime Chevalier-Boisvert
|
852476db7c
Finished renaming MiniGrid methods for PEP8 conformance
|
il y a 7 ans |
|
Maxime Chevalier-Boisvert
|
bfd0f76513
Faster visibility algorithm. Method renamings.
|
il y a 7 ans |
|
Maxime Chevalier-Boisvert
|
15e83a570a
Made gen_obs a public method, renamed public methods.
|
il y a 7 ans |
|
Maxime Chevalier-Boisvert
|
06676a4c74
Fixed issue with agent_sees introduced by visibility changes
|
il y a 7 ans |
