Releases: PathmindAI/nativerl
Releases · PathmindAI/nativerl
Release list
v1.8.1.1 - Hotfix - Fix metrics bug
v1.8.1.0
This is the first release bumping the version number this way.
What's Changed
- Bump version to NativeRL 1.8.1 by @slinlee in #427
- Release Reward terms to Prod by @slinlee in #430
- Revert "Merge pull request #430 from SkymindIO/staging" by @slinlee in #432
- Revert "Merge pull request #432 from SkymindIO/sl/rollbackprod172" by @slinlee in #435
- python 3.7 compat by @maxpumperla in #438
- Update model analyzer build readme by @slinlee in #440
- Add auto-tests and code linting checks for PR by @slinlee in #439
- Re-add createEnvironment from pynativerl by @slinlee in #447
- Fix lint errors in factory sim by @slinlee in #448
- Add description of usage to Readme by @slinlee in #449
- Fix undefined metrics bug #441 by @slinlee in #451
- Weekly update for staging by @slinlee in #452
- Hotfix - Sort base dependencies in reverse order (#456) by @slinlee in #465
- HOTFIX - Remove reverse sorting for libraries by @slinlee in #476
Full Changelog: v1.8.0...v1.8.1.0
v1.8.0.2 - Hotfix - Revert previous hotfix
What's Changed
We modified baseEnv.zip to resolve the classpath conflict. Details: https://3.basecamp.com/3684163/buckets/11875773/uploads/2084974688#__recording_4294816377
Full Changelog: v1.8.0.1...v1.8.0.2
v1.8.0.1 - Hotfix - Sort base dependencies in reverse order
What's Changed
Full Changelog: v1.8.0...v1.8.0.1
v1.7.0
What's Changed
- release dev's change to prod: nativerl 1.3 by @kepricon in #183
- release dev's changes to prod by @kepricon in #213
- release dev's chagnes to prod:v1.4.0 by @kepricon in #233
- release dev's changes to prod by @kepricon in #249
- release Dev's change to Prod by @kepricon in #274
- Hotfix - update AL base env for 8.7.3 (#291) by @slinlee in #293
- Hotfix - fix ecr issue (#280) by @slinlee in #294
- enable PB2 scheduler by @kepricon in #319
- Dev-323 generate git.properties file during build process by @kepricon in #324
- Expose action masking in Pathmind Helper by @ejunprung in #322
- Update Ray configuration by @ejunprung in #320
- Ray 1.3 work by @kepricon in #330
- support longer training by @kepricon in #331
- NN conf configurable by @kepricon in #332
- store averaged metrics across all agent by @kepricon in #335
- change any logic version to 8.7.5 by @kepricon in #337
- Dev-340 disable policy helper for MA by @kepricon in #341
- Support tuple action masking by @ejunprung in #346
- add action masking length check in MA by @kepricon in #354
- use automatically updated library from webapp by @kepricon in #356
- Dev-336 query policy server by @kepricon in #358
Full Changelog: v1.6.0...v1.7.0
Reward terms + Auto-balancing
What's Changed
- release dev's change to prod: nativerl 1.3 by @kepricon in #183
- release dev's changes to prod by @kepricon in #213
- release dev's chagnes to prod:v1.4.0 by @kepricon in #233
- release dev's changes to prod by @kepricon in #249
- release Dev's change to Prod by @kepricon in #274
- Hotfix - update AL base env for 8.7.3 (#291) by @slinlee in #293
- Hotfix - fix ecr issue (#280) by @slinlee in #294
- enable PB2 scheduler by @kepricon in #319
- Dev-323 generate git.properties file during build process by @kepricon in #324
- Expose action masking in Pathmind Helper by @ejunprung in #322
- Update Ray configuration by @ejunprung in #320
- Ray 1.3 work by @kepricon in #330
- support longer training by @kepricon in #331
- NN conf configurable by @kepricon in #332
- store averaged metrics across all agent by @kepricon in #335
- change any logic version to 8.7.5 by @kepricon in #337
- Dev-340 disable policy helper for MA by @kepricon in #341
- Support tuple action masking by @ejunprung in #346
- add action masking length check in MA by @kepricon in #354
- use automatically updated library from webapp by @kepricon in #356
- Dev-336 query policy server by @kepricon in #358
- dev-360 simulation parameter by @kepricon in #362
- dev-360 support more types by @kepricon in #369
- add gamma settings by @kepricon in #373
- support batch mode and rollout fragment length configuration by @kepricon in #375
- add train batch size by @kepricon in #378
- masking + freezing bug fix by @brettskymind in #374
- using object instead stdout by @kepricon in #381
- Save best policy as training continues by @slinlee in #383
- Update Dockerfile by @slinlee in #390
- Make Docker builds for linux bin only by @slinlee in #393
- Add reward balancing by @brettskymind in #372
- Fix escape by @xneyder in #394
- remove rewardSnippet and REWARD_SNIPPET by @alexamakarov in #398
- Update Model Analyzer to use AL 8.7.7 by @slinlee in #401
- Add staging branch for nativerl by @slinlee in #400
- Build and copy nativerl to S3 in Jenkins by @slinlee in #404
- remove conflicting dependencies of Jackson by @alexamakarov in #410
- quick fix to support Py model by @kepricon in #411
- Update staging with latest by @slinlee in #414
- Build model anaylzer docker using training-static S3 bucket by @slinlee in #418
- fix MA by @kepricon in #420
- Update staging with MA fix by @slinlee in #422
- Split ma to components by @alexamakarov in #412
- Fix Pathmind Simulation by @kepricon in #425
- Bump version to 1.8.0 by @slinlee in #426
- Update staging with 1.8.0 by @slinlee in #429
New Contributors
- @brettskymind made their first contribution in #374
Full Changelog: v1.6.0...v1.8.0
v1.7.2 - Add reward balancing
What's Changed
- Add reward balancing by @brettskymind in #372
- Fix escape by @xneyder in #394
- remove rewardSnippet and REWARD_SNIPPET by @alexamakarov in #398
- Update Model Analyzer to use AL 8.7.7 by @slinlee in #401
- Add staging branch for nativerl by @slinlee in #400
- Build and copy nativerl to S3 in Jenkins by @slinlee in #404
- remove conflicting dependencies of Jackson by @alexamakarov in #410
- quick fix to support Py model by @kepricon in #411
- Build model anaylzer docker using training-static S3 bucket by @slinlee in #418
- fix MA by @kepricon in #420
- Split ma to components by @alexamakarov in #412
- Fix Pathmind Simulation by @kepricon in #425
Full Changelog: v1.7.1...v1.7.2
v1.7.1
What's Changed
- dev-360 simulation parameter by @kepricon in #362
- dev-360 support more types by @kepricon in #369
- add gamma settings by @kepricon in #373
- support batch mode and rollout fragment length configuration by @kepricon in #375
- add train batch size by @kepricon in #378
- masking + freezing bug fix by @brettskymind in #374
- using object instead stdout by @kepricon in #381
- Save best policy as training continues by @slinlee in #383
- Update Dockerfile by @slinlee in #390
- Make Docker builds for linux bin only by @slinlee in #393
Full Changelog: v1.7.0...v1.7.1
Release 1.6 - Support AnyLogic 8.7.4 and Ray 1.3
What's Changed
- enable freezing with Webapp by @kepricon in #272
- fix ecr issue by @xneyder in #280
- write complete report after freezing by @kepricon in #282
- Pathmind Helper small bug fixes by @ejunprung in #275
- Python simulations (obs selection, reward functions etc.) by @maxpumperla in #276
- dev-286 fix AL env creation by @kepricon in #287
- update AL base env for 8.7.3 by @kepricon in #291
- modify for mc rollout by @kepricon in #290
- Corrected order of execution of code in step by @SaharEs in #288
- dev-302 support multi agent for freezing by @kepricon in #303
- Do not trigger actions when all agents skipped by @ejunprung in #306
- cut pathmind api by @maxpumperla in #297
- support freezing is false by @kepricon in #307
- support AL 8.7.4 by @kepricon in #305
- Update stopper for Ray 1.3 by @ejunprung in #310
Full Changelog: v1.5.0...v1.6.0