Roman Dobosz
7cbbb8f718
Removed carla environ
2018-05-10 09:39:29 +02:00
Roman Dobosz
cd6376f821
removing doom env
2018-05-10 09:19:32 +02:00
Roman Dobosz
5d47368972
Celaning up coach code + removing play/Human agent
2018-05-10 09:07:24 +02:00
Roman Dobosz
50d38b4b98
Moved main module to cli
2018-05-08 11:01:57 +02:00
Roman Dobosz
26a2f94f43
Added pyyaml dependecy to setup/requirements
2018-04-27 14:48:43 +02:00
Roman Dobosz
676c69e391
Moved coach to its top level module.
2018-04-27 13:25:58 +02:00
Roman Dobosz
7e61bb5685
Removed unnecessary files
2018-04-25 11:59:05 +02:00
Roman Dobosz
5c53f9be02
Added missing imports, correct usages
2018-04-24 13:33:10 +02:00
Roman Dobosz
42a9ec132d
Merge branch 'master' into imports
2018-04-24 07:43:04 +02:00
Itai Caspi
f31159aad6
bug fixes for carla environment ( #93 )
2018-04-23 11:13:24 +03:00
Itai Caspi
52eb159f69
multiple bug fixes in dealing with measurements + CartPole_DFP preset ( #92 )
2018-04-23 10:44:46 +03:00
itaicaspi-intel
5d5562bf62
moving the docs to github
2018-04-23 09:14:20 +03:00
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
jtoy
cafa152382
update requirements to have valid tornado version ( #84 )
2018-04-02 14:21:35 +03:00
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
eba900067c
remove debug
2018-02-21 10:05:57 -05:00
Zach Dwiel
d1bf83047c
remove debug
2018-02-21 10:05:57 -05:00
Zach Dwiel
ef46e194af
remove unused commented code
2018-02-21 10:05:57 -05:00
Zach Dwiel
d9303e731e
remove python2 compatibility
2018-02-21 10:05:57 -05:00
Zach Dwiel
ec68bd4959
make sure that for now observation spaces all include an observation key
2018-02-21 10:05:57 -05:00
Zach Dwiel
0740ebcdac
by default assume state["observation"] is where the image for rendering can be found
2018-02-21 10:05:57 -05:00
Zach Dwiel
f9f92a42fd
cleanup debugging code
2018-02-21 10:05:57 -05:00
Zach Dwiel
86362683b1
comment
2018-02-21 10:05:57 -05:00
Zach Dwiel
8fc24a2bbe
fix bc_agent
2018-02-21 10:05:57 -05:00
Zach Dwiel
d8f5a35013
fix qr_dqn_agent
2018-02-21 10:05:57 -05:00
Zach Dwiel
e1ad86417f
fix n_step_q_agent
2018-02-21 10:05:57 -05:00
Zach Dwiel
5cf10e5f52
fix bug in ddpg
2018-02-21 10:05:57 -05:00
Zach Dwiel
8248caf35e
fix more agents
2018-02-21 10:05:57 -05:00
Zach Dwiel
98f57a0d87
fix ddpg
2018-02-21 10:05:57 -05:00
Zach Dwiel
943e41ba58
fix nec_agent
2018-02-21 10:05:57 -05:00
Zach Dwiel
ee6e0bdc3b
fix keep_dims -> keepdims
2018-02-21 10:05:57 -05:00
Zach Dwiel
39a28aba95
fix clipped ppo
2018-02-21 10:05:57 -05:00
Zach Dwiel
85afb86893
temp commit
2018-02-21 10:05:57 -05:00
Gal Leibovich
16c5032735
fix for tensorboard visualization slowing execution even when it is off
...
apparently tensorflow still collect summary data even when no summary FileWriter is defined.
2018-02-18 16:35:24 +02:00
Itai Caspi
72d34f4063
adding a flag to prevent summary
2018-02-15 13:47:14 +02:00
Itai Caspi
55c8c87afc
allow visualizing the observation + bug fixes to coach summary
2018-02-15 13:47:14 +02:00
Itai Caspi
5d1a2bc392
Adding a summary when exiting coach
2018-02-13 11:11:26 +02:00
Itai Caspi
ba96e585d2
appending csv's from logger instead of rewriting them
2018-02-12 14:52:50 +02:00
Itai Caspi
569ca39ce6
Dashboard color selection + removing old legend
2018-02-09 14:52:58 +02:00
Itai Caspi
8a4383e86f
Added an improved legend to dashboard
2018-02-08 16:48:46 +02:00
Itai Caspi
b071599cb0
updating intel optimized tensorflow to version 1.4
2018-02-08 09:20:29 +02:00
Itai Caspi
462fe9796b
several bug fixes in dashboard
2018-02-07 12:49:34 +02:00
galleibo-intel
4025496783
Setting tensorflow-gpu version to 1.4.1 (1.5.0 is not tested yet)
2018-02-05 15:48:00 +02:00
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
Itai Caspi
a8d5fb7bdf
Added a table of contents to the README
2018-01-27 14:31:53 +02:00
Itai Caspi
522c837e76
Update README.md
2018-01-22 12:15:23 +02:00
Itai Caspi
43821c9630
adding the selu activation
2018-01-22 12:05:43 +02:00
Zach Dwiel
fff8c8f568
provide a helpful error message in the event that an exploration policy returns a vector of actions instead of a single action during value optimization agent
2018-01-20 14:11:24 -05:00
Zach Dwiel
40e5c628c6
add options for more verbose test errors
2018-01-16 22:08:46 -05:00
