Commit Graph

71 Commits

Author SHA1 Message Date
Philipp Horstenkamp 5addc1ffa1 Update README.md 2023-10-26 01:46:11 +02:00
Philipp Horstenkamp b603a775dc Update README.md 2023-10-26 01:36:19 +02:00
Philipp Horstenkamp 74a6f429b3 Update README.md 2023-10-26 01:34:54 +02:00
Philipp Horstenkamp 1d0fc1a6a2
Removed old training data. 2023-03-31 22:36:38 +02:00
Philipp Horstenkamp 742ac0a4e1
Final verion up. 2023-03-31 22:29:56 +02:00
Philipp Horstenkamp 2f64f17026
First draft. 2023-03-31 21:58:54 +02:00
Philipp Horstenkamp be94f11274
Lots of text and analysis added 2023-03-31 19:08:58 +02:00
Philipp Horstenkamp 791a3c694e
Added some first evaluations. 2023-03-31 11:43:44 +02:00
Philipp Horstenkamp 078463a4de
Lots of changes 2023-03-31 02:12:23 +02:00
Philipp Horstenkamp c923a26ede
Lots of changes 2023-03-31 00:26:39 +02:00
Philipp Horstenkamp a18cf0beb6
Added most of the lost text back in again. 2023-03-31 00:15:27 +02:00
Philipp Horstenkamp 25187122d8
Some updates 2023-03-30 23:54:37 +02:00
Philipp Horstenkamp dcf8aae87e
Some updates 2023-03-30 23:54:26 +02:00
Philipp Horstenkamp d94f83c9e3
Merge remote-tracking branch 'origin/main' into main
# Conflicts:
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-10.pickle
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-10.torch
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-11.pickle
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-11.torch
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-12.pickle
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-12.torch
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-13.pickle
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-13.torch
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-14.pickle
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-14.torch
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-15.pickle
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-15.torch
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-16.pickle
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-16.torch
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-17.pickle
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-17.torch
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-18.pickle
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-18.torch
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-19.pickle
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-19.torch
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-20.pickle
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-20.torch
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-21.pickle
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-21.torch
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-22.pickle
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-22.torch
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-23.pickle
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-23.torch
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-24.pickle
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-24.torch
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-25.pickle
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-25.torch
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-3.pickle
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-3.torch
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-4.pickle
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-4.torch
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-5.pickle
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-5.torch
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-6.pickle
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-6.torch
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-7.pickle
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-7.torch
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-8.pickle
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-8.torch
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-9.pickle
#	training_data/QL-M-G08-WW00-FSF10-DQLSimple-MSELoss-9.torch
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-10.pickle
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-10.torch
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-11.pickle
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-11.torch
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-12.pickle
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-12.torch
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-13.pickle
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-13.torch
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-14.pickle
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-14.torch
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-15.pickle
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-15.torch
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-16.pickle
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-16.torch
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-17.pickle
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-17.torch
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-18.pickle
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-18.torch
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-19.pickle
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-19.torch
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-20.pickle
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-20.torch
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-21.pickle
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-21.torch
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-3.pickle
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-3.torch
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-4.pickle
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-4.torch
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-5.pickle
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-5.torch
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-6.pickle
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-6.torch
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-7.pickle
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-7.torch
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-8.pickle
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-8.torch
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-9.pickle
#	training_data/QL-M-G09-WW00-FSF10-DQLSimple-MSELoss-9.torch
2023-03-30 23:34:33 +02:00
Philipp Horstenkamp d75d290eb7
Lots of training done 2023-03-30 23:33:31 +02:00
Philipp Horstenkamp f9b0845907 Added lots of training. 2023-03-30 23:18:07 +02:00
Philipp Horstenkamp d9528eb5a1
Reomved defekt training 2023-03-30 02:37:58 +02:00
Philipp Horstenkamp 251376678b
Added a lot of training history 2023-03-30 02:29:34 +02:00
Philipp Horstenkamp 2229f7e416
Some more warnings corrected 2023-03-06 22:44:49 +01:00
Philipp Horstenkamp 338289b664
Added a lot of training data 2023-03-06 22:41:01 +01:00
Philipp Horstenkamp e9cad87c57
Repaired a lot of smaller things that where not quit smooth 2023-03-06 22:37:51 +01:00
Philipp Horstenkamp b9df1b3093
Lots of small documentation additions. 2023-03-06 00:11:35 +01:00
Philipp Horstenkamp 6bec4b941c
Lots of changes 2023-02-28 23:48:13 +01:00
Philipp Horstenkamp cdbd6dc4be
Fixed the order player play the game 2023-02-26 16:06:12 +01:00
Philipp Horstenkamp ef122fb6d0
Repaired a bug in the q learning reword function 2023-02-22 21:55:08 +01:00
Philipp Horstenkamp 138d9df2b9
Added hash branch exploration 2023-02-22 21:36:43 +01:00
Philipp Horstenkamp 3bd848f92d
Change the exploration to switch on an alternative 2023-02-22 02:54:52 +01:00
Philipp Horstenkamp 1428768a5c
Bugfixes 2023-02-22 02:15:16 +01:00
Philipp Horstenkamp 1510d7fa4d
Some more bugfixes
Added a few more bugfixes
2023-02-22 02:14:16 +01:00
Philipp Horstenkamp d66c951a0d
Fixed some bugs 2023-02-21 23:02:32 +01:00
Philipp Horstenkamp cfa8e4226f
Fixed lots of typing errors and typos 2023-02-20 19:25:27 +01:00
Philipp Horstenkamp 8b2b0a2efb
Update main 2023-02-20 18:27:03 +01:00
Philipp Horstenkamp 4e16511568
Added a frist train function
Added a first training function
2023-02-19 23:40:11 +01:00
Philipp Horstenkamp 464fa2e419
Added a config file to the gitignore 2023-02-19 03:22:25 +01:00
Philipp Horstenkamp 9cf9fe820e
Added the ability to generate training data. 2023-02-19 03:18:48 +01:00
Philipp Horstenkamp 3dc5a5d1cf
Poetry update 2023-02-19 00:58:31 +01:00
Philipp Horstenkamp 7cc8b6c025
Added a first network 2023-02-18 23:40:00 +01:00
Philipp Horstenkamp fc65735bca
Poetry update 2023-02-18 23:23:58 +01:00
Philipp Horstenkamp 0c9cf50bc5
Some debugging of the q reword function 2023-02-18 14:33:56 +01:00
Philipp Horstenkamp 80abd61475
Added the analysis of possible turn 2023-02-18 00:21:26 +01:00
Philipp Horstenkamp c0943e4309
Added the points per score at turn label. 2023-02-18 00:12:29 +01:00
Philipp Horstenkamp e199c9ab55
Reworked some plots 2023-02-18 00:03:13 +01:00
Philipp Horstenkamp dfe3b3aa59
Added a greedy policy 2023-02-17 03:04:19 +01:00
Philipp Horstenkamp ef20f3f68a
Added a statistical example of the game. 2023-02-17 02:16:24 +01:00
Philipp Horstenkamp c33c5931f3
Added npy files to LFS config. 2023-02-17 01:47:49 +01:00
Philipp Horstenkamp aa7eb02389
added kdepy and plotly 2023-02-17 01:47:18 +01:00
Philipp Horstenkamp ea2180b08b
Added statistical analysis 2023-02-16 02:18:04 +01:00
Philipp Horstenkamp 0c81f2d006
Added a reword function for q learning. 2023-02-13 03:37:00 +01:00
Philipp Horstenkamp 9b011b548e
Fixed a bug in the assignment of invalid turns. Added lots of documentation. 2023-02-13 00:42:25 +01:00
Philipp Horstenkamp 29bbd83467
Uses html instead of md to include images. 2023-02-12 20:40:38 +01:00