Experimental Code #56

bcollazo · 2021-01-29T19:15:46Z

No description provided.

…mps), docker prototyping

coveralls · 2021-01-29T19:36:03Z

Pull Request Test Coverage Report for Build 521810993

385 of 413 (93.22%) changed or added relevant lines in 11 files are covered.
9 unchanged lines in 3 files lost coverage.
Overall coverage increased (+0.9%) to 92.165%

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
catanatron/models/player.py	20	21	95.24%
catanatron/models/board.py	148	153	96.73%
catanatron/game.py	119	141	84.4%

Files with Coverage Reduction	New Missed Lines	%
catanatron/game.py	1	85.67%
catanatron/models/board.py	2	95.26%
catanatron/models/actions.py	6	92.81%

Totals
Change from base Build 376572224:	0.9%
Covered Lines:	1094
Relevant Lines:	1187

💛 - Coveralls

bcollazo added 30 commits November 21, 2020 23:03

PREDICTING ACTIONS INCOMPLETE

9317d9d

Be robust to seating order in ML Features

2c0a0d6

COMMITABLE: Cleanup

a9fc409

COMMITABLE: state, action on replay

1a80dc0

Forward-computing samples. and experiment working

042e36f

Add logs to .gitignore

68f21f1

ML Progress

91e6a8e

ML: better .gitignore

e93ed53

Assets Prototyping

a48e2ca

ML: Correct features (hide enemy dev playable)

cd098ed

ML: Server expose sample per player

ee894b1

ML: Qlearn cleanup

f9f9060

ML: Try different returns, change model loc

b02446a

ML: Fix greedy_estimate player and update models

45167c0

ML: Faster QRLPlayer

c8d89af

ML: Leanify qlearn.py

ecd3225

ML: Model cleanup

18a5f25

ML: VRLPlayer wins 60% on random data. Generalized train.py

84f745a

ML+Core: Player.restart_state(). Prep for model versioning

21fd7e8

ML: VRLPlayer v1,v2... folders.

3057a82

ML: Versioning models (training too).

7f23831

ML: Miscellaneous README.md

7f4f606

ML: Eliminate GreedEstimatePlayer

74bd8c1

ML: Small improvement. Make model serializable.

31f61d5

New tile and remove UI warning

96051c1

Add pytest-benchmark

a7c3f48

pytest-benchmark

3ef9a21

pytest-benchmark dont autosave

3d4a022

cleanup creat_feature-matrix.py

214c893

fix pytest-benchmark

86c3366

bcollazo added 27 commits December 20, 2020 16:55

Update model. 80% vmodel

1451c48

setup pseudoimage ml dimensions

6015f8c

Cleanup

bb70813

coordinate_system.py conversions

6790b30

comment improvement

d03c309

Board Tensor representation v1

539e0bc

bryan.py changes

4c6aaf2

save tensor board when playing

fe2c357

scripts folder

e0c7595

Play with board tensor player

b7f2089

normalized model (saved with pydot). gzip compress

26fca9d

Initial tensor board model

d0c9817

More TensorBoard Model V1 material.

90fd43b

Play server with models

bf6fb08

Move modesl to models folder

2ce423a

Experimental Folder, fastest python game impl yet (cache connected co…

f52b9e8

…mps), docker prototyping

Improvements to Docker. Not quite there yet

7e7d18f

Move to primitive type Actions. Fix bugs.

cec0b40

MCTS player, best one yet.

664d9ed

All-in-one Docker working

ff4aab3

Scripts

6aedb55

Bug fixes. Including tensor representation

1ea4ffc

small cleanup

ad91b9e

Some files modified

9d35ef4

Removing gitattributes

505889a

restore files from lfs:

5d5d0a5

attempt to fix coveralls

d65f2fe

bcollazo merged commit 9f96e2d into master Jan 29, 2021

bcollazo deleted the bryan/fastest-docker-mctsv1 branch January 29, 2021 19:36

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Experimental Code #56

Experimental Code #56

bcollazo commented Jan 29, 2021

coveralls commented Jan 29, 2021

Experimental Code #56

Experimental Code #56

Conversation

bcollazo commented Jan 29, 2021

coveralls commented Jan 29, 2021

Pull Request Test Coverage Report for Build 521810993

💛 - Coveralls