Skip to content

Conversation

andrewcoh
Copy link
Contributor

@andrewcoh andrewcoh commented Dec 3, 2020

Proposed change(s)

This was non trivial due to the clipping changes. Also, there seems to be many Match 3 changes causing a large diff

Useful links (Github issues, JIRA tickets, ML-Agents forum threads etc.)

Types of change(s)

  • Bug fix
  • New feature
  • Code refactor
  • Breaking change
  • Documentation update
  • Other (please describe)

Checklist

  • Added tests that prove my fix is effective or that my feature works
  • Updated the changelog (if applicable)
  • Updated the documentation (if applicable)
  • Updated the migration guide (if applicable)

Other comments

Hunter-Unity and others added 30 commits November 9, 2020 16:12
* first pass - matching and gravity work

* valid moves WIP

* move enumeration, simple checking

* move detection

* fix up/down directions

* actuator and heuristic

* sensor

* reset if no moves

* add masking, retrain

* cleanup

* config update

* optionally force heuristic mode

* compressed observations, reset board if no moves

* TFModel meta files

* WIP abstract

* random move heuristic

* rename Move fields, migrate IsValid logic

* comments

* reduce dependency on Match3Agent

* abstractBoard as monobehavior

* move actuator to extensions

* move sensor to extensions

* update components in scene

* undo accidental checkins

* drawing cleanup, custom maxMoves logic

* meta file

* prefab, use 8 agents

* different behaviors

* timers, early out

* unit tests

* docs

* Apply suggestions from code review

Co-authored-by: Vincent-Pierre BERGES <[email protected]>

* doc cleanup, swap to .nn file, add heuristic agents

* board iterator

* convert others to iterator, rename to Move.MoveIndex

* change observation type to enum

* fix flip for compressed obs

* save dimensions on sensor

* save dimension on actuator

* Match3 example - special pieces (#4575)

* WIP special pieces

* add special piece types to obs (need more from master first)

* update sensors and models

* docstrings

* more interesting scoring, smarter heuristic

* more steps

* longer trained models

* handle no valid moves more gracefully (#4598)

* handle no valid moves more gracefully

* rename callback

* cleanup heuristic logic

* more sensor and actuator tests

* update heuristic configs

* docstrings and add to environment docs

* sensor and actuator names

* compare compressed obs against saved PNGs

* review feedback, cleanup TODOs, comments

* docstrings, heuristic benchmarks

* kind of working

* added symbol gameobjects

* basic shapes & colors working

* cleanup and added background

* bug hunt. won't train

* merge master version

* update prefabs

* update prefabs

* add image to docs

* Update Match3.md

* rename some properties

* remove unneeded comments and some cleanup

Co-authored-by: Chris Elion <[email protected]>
Co-authored-by: Vincent-Pierre BERGES <[email protected]>
* Update CONTRIBUTING.md

Added note about env contributions

* removed trailing space
…ile (#4630)

* Update Training-ML-Agents.md

* Update docs/Training-ML-Agents.md

Co-authored-by: Ervin T. <[email protected]>

* removed trailing whitespace

Co-authored-by: Ervin T. <[email protected]>
* Change title

* added overvierw, cleaned up structure.

* Updated with getting started, clean up

* added link to form.  added note about playable

* clean up

* Update Match3.md
* use int64 steps

* check for NaN actions

Co-authored-by: Ruo-Ping Dong <[email protected]>
* Update Readme table and loose link to master

* Remove ../.. and replace with release_10_docs
* use int64 steps

* check for NaN actions

Co-authored-by: Ruo-Ping Dong <[email protected]>

Co-authored-by: Chris Elion <[email protected]>
* updated image in learning envs examples

* add link to learning example to match-3

* cleaned up headings

* removed anchor

* Update Match3.md

* Delete match3.png

* added new match3 image

* updated match3 image link
…ing Match3 (#4664)

* match3 settings

* Add epsilon to log

* Add another epsilon

* Revert match3 configs

* NaN-free masking method

* Add comment for paper

* Add comment for paper

Co-authored-by: Chris Elion <[email protected]>
* match3 settings

* Add epsilon to log

* Add another epsilon

* Revert match3 configs

* NaN-free masking method

* Add comment for paper

* Add comment for paper

Co-authored-by: Chris Elion <[email protected]>

Co-authored-by: Chris Elion <[email protected]>
* [Bug-fix] Remove extra directory created at checkpoint

* removing os import
* Remove extra period

* Update changelog

* Re add a period (#4676)

* Adding a trailing period

* formating

Co-authored-by: Vincent-Pierre BERGES <[email protected]>
* [Bug-fix] Remove extra directory created at checkpoint

* removing os import
Ervin T and others added 8 commits November 19, 2020 17:26
* Remove extra period

* Update changelog

* Re add a period (#4676)

* Adding a trailing period

* formating

Co-authored-by: Vincent-Pierre BERGES <[email protected]>

Co-authored-by: Vincent-Pierre BERGES <[email protected]>
* pin pip to 20.2 for now

* add self to list of trigger files
@andrewcoh andrewcoh merged commit acc9bed into develop-hybrid-action-staging Dec 3, 2020
@delete-merged-branch delete-merged-branch bot deleted the develop-staging-merge-master branch December 3, 2020 23:55
@github-actions github-actions bot locked as resolved and limited conversation to collaborators Dec 4, 2021
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

5 participants