7 Commits

Author	SHA1	Message	Date
Roman Dobosz	1b095aeeca	Cleanup imports. ... Till now, most of the modules were importing all of the module objects (variables, classes, functions, other imports) into module namespace, which potentially could (and was) cause of unintentional use of class or methods, which was indirect imported. With this patch, all the star imports were substituted with top-level module, which provides desired class or function. Besides, all imports where sorted (where possible) in a way pep8[1] suggests - first are imports from standard library, than goes third party imports (like numpy, tensorflow etc) and finally coach modules. All of those sections are separated by one empty line. [1] https://www.python.org/dev/peps/pep-0008/#imports	2018-04-13 09:58:40 +02:00
Itai Caspi	a7206ed702	Multiple improvements and bug fixes (#66) ... * Multiple improvements and bug fixes: * Using lazy stacking to save on memory when using a replay buffer * Remove step counting for evaluation episodes * Reset game between heatup and training * Major bug fixes in NEC (is reproducing the paper results for pong now) * Image input rescaling to 0-1 is now optional * Change the terminal title to be the experiment name * Observation cropping for atari is now optional * Added random number of noop actions for gym to match the dqn paper * Fixed a bug where the evaluation episodes won't start with the max possible ale lives * Added a script for plotting the results of an experiment over all the atari games	2018-02-26 12:29:07 +02:00
Zach Dwiel	4fe9cba445	remove debug	2018-02-21 10:05:57 -05:00
Zach Dwiel	85afb86893	temp commit	2018-02-21 10:05:57 -05:00
Gal Leibovich	7c8962c991	adding support in tensorboard (#52) ... * bug-fix in architecture.py where additional fetches would acquire more entries than it should * change in run_test to allow ignoring some test(s)	2018-02-05 15:21:49 +02:00
Zach Dwiel	40e5c628c6	add options for more verbose test errors	2018-01-16 22:08:46 -05:00
Itai Caspi	125c7ee38d	Release 0.9 ... Main changes are detailed below: New features - * CARLA 0.7 simulator integration * Human control of the game play * Recording of human game play and storing / loading the replay buffer * Behavioral cloning agent and presets * Golden tests for several presets * Selecting between deep / shallow image embedders * Rendering through pygame (with some boost in performance) API changes - * Improved environment wrapper API * Added an evaluate flag to allow convenient evaluation of existing checkpoints * Improve frameskip definition in Gym Bug fixes - * Fixed loading of checkpoints for agents with more than one network * Fixed the N Step Q learning agent python3 compatibility	2017-12-19 19:27:16 +02:00