sawcordwell/pymdptoolbox issues and pull requests

#44 - Recycling robot MDP

Issue - State: closed - Opened by JreigeF 7 months ago

#43 - Not working when initial_value is numpy.

Pull Request - State: open - Opened by Al17OTON over 1 year ago

#43 - Not working when initial_value is numpy.

Pull Request - State: open - Opened by Al17OTON over 1 year ago

#42 - Please release a new version with the latest changes

Issue - State: open - Opened by hemanik about 2 years ago

#41 - Fix typo in doc

Pull Request - State: open - Opened by yetanotherportfolio about 2 years ago

#40 - Second test of git branch

Pull Request - State: closed - Opened by dhuberdeau over 2 years ago

#39 - Why the values of policy iteration and the values of value iteration are different?

Issue - State: open - Opened by leelening almost 3 years ago

#38 - Update mdp.py

Pull Request - State: closed - Opened by strgalt-t over 3 years ago

#37 - Question about undiscounted model

Issue - State: open - Opened by EfthymiaKostaki almost 4 years ago

#36 - thanks

Issue - State: open - Opened by azs1997421 almost 4 years ago

#35 - Linear Programming algo

Issue - State: open - Opened by glarange about 4 years ago

#34 - Why changed the epsilon in Q-learning and the way to update Q, is this better?

Issue - State: open - Opened by baimengwei over 4 years ago

#33 - Stopping Criteria || pymdptoolbox || boundIter() function

Issue - State: open - Opened by 3shmawei almost 5 years ago

#32 - Fix division by zero in `ValueIteration._boundIter`

Pull Request - State: open - Opened by AdamGleave almost 5 years ago

#31 - Add ability to use custom learning rate

Pull Request - State: open - Opened by JSPromisel over 5 years ago - 1 comment

#30 - User guide?

Issue - State: open - Opened by birdybird almost 6 years ago - 1 comment

#29 - Improper Assertion Statement.

Issue - State: open - Opened by ryanpeach about 6 years ago

#28 - Fix capitalization of "Q-learning".

Pull Request - State: open - Opened by crypto-jeronimo about 6 years ago - 1 comment

#27 - Fix broken examples

Pull Request - State: open - Opened by crypto-jeronimo about 6 years ago - 1 comment

#26 - how to train it?

Issue - State: open - Opened by JimmyCXXQ over 6 years ago

#25 - MDP where not all actions are always available

Issue - State: open - Opened by jniediek over 6 years ago - 4 comments

#24 - Solution for basic grid world example

Issue - State: open - Opened by teldridge11 over 6 years ago - 1 comment

#23 - #Fix max_iter parameter setting.

Pull Request - State: closed - Opened by BaptisteAmato over 6 years ago

#22 - Switched broken pypip.in badges to shields.io

Pull Request - State: open - Opened by movermeyer over 6 years ago - 1 comment

#21 - Absolute difference is unnecessarily space-inefficient

Issue - State: open - Opened by bissias over 6 years ago - 1 comment

#20 - skip_check

Issue - State: open - Opened by dlamghariidrissi about 7 years ago

#19 - Model-free algorithms depend on model

Issue - State: open - Opened by sovelten about 8 years ago - 3 comments

#18 - pip install issues

Issue - State: open - Opened by onaclov2000 about 9 years ago - 1 comment

#17 - MDP solving with LP

Pull Request - State: open - Opened by silgon about 9 years ago - 2 comments

#16 - Numpy Version

Issue - State: open - Opened by musicarroll about 9 years ago

#15 - Tests for undiscounted MDPs

Pull Request - State: open - Opened by yasserglez over 9 years ago

#14 - ValueIterationGS _boundIter is incorrect

Issue - State: open - Opened by sawcordwell over 9 years ago

#13 - max_iteration parameters is not respected in the ValueIteration Algorithm

Issue - State: open - Opened by sshegheva over 9 years ago - 3 comments

#12 - fixed firemdp.py AssertionError caused by incorrect argument sequence

Pull Request - State: closed - Opened by thejat over 9 years ago - 2 comments

#11 - Add skip_check argument in MDP and subclasses

Pull Request - State: closed - Opened by yasserglez almost 10 years ago - 2 comments

#10 - Sparse rewards are converted to dense arrays

Issue - State: open - Opened by sawcordwell almost 10 years ago - 1 comment

#9 - Linear programming class is broken

Issue - State: open - Opened by sawcordwell almost 10 years ago - 1 comment
Labels: bug

#8 - WIP: [mdp] Import the util module as _util

Pull Request - State: closed - Opened by sawcordwell almost 10 years ago - 1 comment

#7 - Errors with sparse transition and/or reward matrices

Issue - State: closed - Opened by yasserglez almost 10 years ago - 7 comments
Labels: bug

#6 - Unit tests for undiscounted MDPs required

Issue - State: open - Opened by sawcordwell almost 10 years ago - 2 comments

#5 - Fix V initialization in PolicyIterationModified

Pull Request - State: closed - Opened by yasserglez almost 10 years ago - 2 comments

#4 - [testing] Remove unused travis deps

Pull Request - State: closed - Opened by sawcordwell almost 10 years ago

#3 - Implement own exception class

Issue - State: closed - Opened by sawcordwell almost 10 years ago
Labels: enhancement

#2 - Move evalPolicy* functions to be functions of the util module

Issue - State: open - Opened by sawcordwell almost 10 years ago
Labels: enhancement

#1 - Make sure that doctests are run when testing on Travis CI

Issue - State: closed - Opened by sawcordwell almost 10 years ago
Labels: bug, enhancement

GitHub / sawcordwell/pymdptoolbox issues and pull requests