 Maxime Chevalier-Boisvert
|
ed662a5597
Implemented mechanism to prevent seeing through walls
|
%!s(int64=6) %!d(string=hai) anos |
|
Maxime Chevalier-Boisvert
|
cdb80a5d30
Started work on Playground-v1 environment for curriculum learning
|
%!s(int64=6) %!d(string=hai) anos |
|
Maxime Chevalier-Boisvert
|
e00764ebf8
Lazily importing visdom
|
%!s(int64=6) %!d(string=hai) anos |
|
Maxime Chevalier-Boisvert
|
340c03a446
Cleaned up and simplified _genGrid functions
|
%!s(int64=6) %!d(string=hai) anos |
|
Maxime Chevalier-Boisvert
|
3096a3b7f4
Updated README
|
%!s(int64=6) %!d(string=hai) anos |
|
Maxime Chevalier-Boisvert
|
0e496fe636
Fixed visdom visualization code
|
%!s(int64=6) %!d(string=hai) anos |
|
Maxime Chevalier-Boisvert
|
e5f35ea056
Randomized agent position in playground environment
|
%!s(int64=6) %!d(string=hai) anos |
|
Maxime Chevalier-Boisvert
|
1e5d26e4c0
Fixed issue with environment seeding
|
%!s(int64=6) %!d(string=hai) anos |
|
Maxime Chevalier-Boisvert
|
c4049b13ed
Added Playground-v0 environment for experiments
|
%!s(int64=6) %!d(string=hai) anos |
|
Maxime Chevalier-Boisvert
|
5a6461ff2e
Added GRU to policy, made model larger.
|
%!s(int64=6) %!d(string=hai) anos |
|
Maxime Chevalier-Boisvert
|
17eba78e33
Fixed pyqt5 package version
|
%!s(int64=6) %!d(string=hai) anos |
|
Maxime Chevalier-Boisvert
|
2cdc42ac43
Modified reward range for fetch environment
|
%!s(int64=6) %!d(string=hai) anos |
|
Maxime Chevalier-Boisvert
|
2b1d180dda
Corrected reward ranges for environments
|
%!s(int64=6) %!d(string=hai) anos |
|
Maxime Chevalier-Boisvert
|
24f7678f57
Update README.md
|
%!s(int64=6) %!d(string=hai) anos |
|
Maxime Chevalier-Boisvert
|
27e33995ab
Fixed wrappers.py following changes in OpenAI gym
|
%!s(int64=6) %!d(string=hai) anos |
|
Maxime Chevalier-Boisvert
|
25fe4664fa
Modified environments so they all produce observations in a dict
|
%!s(int64=6) %!d(string=hai) anos |
|
Maxime Chevalier-Boisvert
|
7c41bdff21
Added MiniGrid-MultiRoom-N2-S4-v0 environment
|
%!s(int64=6) %!d(string=hai) anos |
|
Maxime Chevalier-Boisvert
|
4d84ecd45f
Eliminated source of non-determinism
|
%!s(int64=6) %!d(string=hai) anos |
|
Maxime Chevalier-Boisvert
|
d7b381ce47
Added "in" operator for Grid objects
|
%!s(int64=6) %!d(string=hai) anos |
|
Maxime Chevalier-Boisvert
|
dcc9732274
Added env.agentSees(x, y) method
|
%!s(int64=6) %!d(string=hai) anos |
|
Maxime Chevalier-Boisvert
|
50bd721381
Removed inaccurate comment
|
%!s(int64=6) %!d(string=hai) anos |
|
Maxime Chevalier-Boisvert
|
bb6dc6196e
Added test
|
%!s(int64=6) %!d(string=hai) anos |
|
Maxime Chevalier-Boisvert
|
114caa944a
Fixes based on changes in OpenAI Gym 0.9.6
|
%!s(int64=6) %!d(string=hai) anos |
|
Maxime Chevalier-Boisvert
|
ea0e67e005
Added environments to README
|
%!s(int64=6) %!d(string=hai) anos |
|
Maxime Chevalier-Boisvert
|
99f583af9e
Completed LockedRoom environment
|
%!s(int64=6) %!d(string=hai) anos |
|
Maxime Chevalier-Boisvert
|
d70e134948
Eliminated WrapPyTorch
|
%!s(int64=6) %!d(string=hai) anos |
|
Maxime Chevalier-Boisvert
|
c46ade2f4f
Fixed bug in fetch environment
|
%!s(int64=6) %!d(string=hai) anos |
|
Maxime Chevalier-Boisvert
|
cd33e57ae6
Updated default arguments for RL code
|
%!s(int64=6) %!d(string=hai) anos |
|
Maxime Chevalier-Boisvert
|
aa65f2f84f
Changed observation_space for putnear
|
%!s(int64=6) %!d(string=hai) anos |
|
Maxime Chevalier-Boisvert
|
16085191ab
Refactored handling of recurrent policies for simplicity
|
%!s(int64=6) %!d(string=hai) anos |
