Commit graph

28 commits

Author SHA1 Message Date
ddelange
e109c13b83
[ci] Clean up ray-ml requirements (#23325)
In https://github.com/ray-project/ray/blob/ray-1.11.0/docker/ray-ml/Dockerfile, the order of pip install commands currently matters (potentially a lot). It would be good to run one big pip install command to avoid ending up with a broken env.

Co-authored-by: Kai Fricke <krfricke@users.noreply.github.com>
2022-03-25 15:59:54 +00:00
Amog Kamsetty
d31d6bc9bb
[Docker] Add Train requirements to ray-ml docker image (#22645) 2022-03-17 15:07:32 -07:00
Kai Fricke
84a163a2c4
[RLlib] Remove atari rom install script (#22797) 2022-03-03 16:55:56 +01:00
Sven Mika
4cb23d1c95
[Tune; Testing] Revert to 3.7 (undone by accident by previous PR); + some minor comment cleanups. (#20031) 2021-11-04 10:58:34 +01:00
Avnish Narayan
026bf01071
[RLlib] Upgrade gym version to 0.21 and deprecate pendulum-v0. (#19535)
* Fix QMix, SAC, and MADDPA too.

* Unpin gym and deprecate pendulum v0

Many tests in rllib depended on pendulum v0,
however in gym 0.21, pendulum v0 was deprecated
in favor of pendulum v1. This may change reward
thresholds, so will have to potentially rerun
all of the pendulum v1 benchmarks, or use another
environment in favor. The same applies to frozen
lake v0 and frozen lake v1

Lastly, all of the RLlib tests and have
been moved to python 3.7

* Add gym installation based on python version.

Pin python<= 3.6 to gym 0.19 due to install
issues with atari roms in gym 0.20

* Reformatting

* Fixing tests

* Move atari-py install conditional to req.txt

* migrate to new ale install method

* Fix QMix, SAC, and MADDPA too.

* Unpin gym and deprecate pendulum v0

Many tests in rllib depended on pendulum v0,
however in gym 0.21, pendulum v0 was deprecated
in favor of pendulum v1. This may change reward
thresholds, so will have to potentially rerun
all of the pendulum v1 benchmarks, or use another
environment in favor. The same applies to frozen
lake v0 and frozen lake v1

Lastly, all of the RLlib tests and have
been moved to python 3.7
* Add gym installation based on python version.

Pin python<= 3.6 to gym 0.19 due to install
issues with atari roms in gym 0.20

Move atari-py install conditional to req.txt

migrate to new ale install method

Make parametric_actions_cartpole return float32 actions/obs

Adding type conversions if obs/actions don't match space

Add utils to make elements match gym space dtypes

Co-authored-by: Jun Gong <jungong@anyscale.com>
Co-authored-by: sven1977 <svenmika1977@gmail.com>
2021-11-03 16:24:00 +01:00
Amog Kamsetty
5d54412f1c
[Docker] Alias ray-ml:nightly to ray-ml:nightly-gpu (#19726)
* wip

* wip

* update

* finish

* deprecate

* debug

* fix and address comments

* try catch

* fix

* split tests

* force

* merge

* docs

* wip

* fix and check

* update readme

* fix

* fix

* fix sanity checking

* format

* alias

* fix

* comment
2021-10-27 11:30:49 -07:00
Amog Kamsetty
db863aafc0
Revert "Revert "[Docker] Support multiple CUDA Versions (#19505)" (#19756)" (#19763)
This reverts commit e58fcca404.
2021-10-26 17:32:56 -07:00
Amog Kamsetty
e58fcca404
Revert "[Docker] Support multiple CUDA Versions (#19505)" (#19756)
This reverts commit f0053d405b.
2021-10-26 12:55:20 -07:00
Amog Kamsetty
f0053d405b
[Docker] Support multiple CUDA Versions (#19505)
* wip

* wip

* update

* finish

* deprecate

* debug

* fix and address comments

* try catch

* fix

* split tests

* force

* merge

* docs

* wip

* fix and check

* update readme

* fix

* fix

* fix sanity checking

* format
2021-10-25 18:57:05 -07:00
Amog Kamsetty
84e958f330
[ML] Consolidate and upgrade Deep Learning Dependencies (#18574)
* wip
'

* upgrade requirements

* add file

* fix

* fixes

* Apply suggestions from code review

Try mlagents==0.21.0 for now (works with torch 1.9).

* Apply suggestions from code review

* wip

* wip

* fix

* fix

* upgrade lightning bolts

* address comment

Co-authored-by: Sven Mika <sven@anyscale.io>
2021-09-16 20:16:40 -07:00
Sven Mika
8a00154038
[RLlib] Bump tf version in ML docker to tf==2.5.0; add tfp to ML-docker. (#18544) 2021-09-15 08:46:37 +02:00
Kai Fricke
fb38d06cfb
Move RLLib GPU release test dependencies to ml docker (#18208) 2021-09-03 09:35:18 +01:00
Amog Kamsetty
544dff80fa
[Docker] Fix torch GPU install on Ray Docker images (#15473)
Co-authored-by: Ian Rodney <ian.rodney@gmail.com>
2021-04-26 16:22:25 -07:00
Richard Liaw
d92c00e233
Pin autogluon.core for builds (#14448) 2021-03-02 15:55:03 -08:00
Richard Liaw
1e25747818
Fix docker builds by marking resolver (#14445)
Signed-off-by: Richard Liaw <rliaw@berkeley.edu>
2021-03-02 09:36:40 -08:00
Amog Kamsetty
a430ac2334
[Tune] Revert Pinning Tune Dependencies (#14059)
* remove lockfiles

* docker

* remove constraint file

* fix
2021-02-11 15:43:09 -08:00
Ian Rodney
813a7ab0e2
[docker] Build Python3.6 & Python3.8 Docker Images (#13548) 2021-01-28 15:24:50 -08:00
Amog Kamsetty
3f42e6bafe
[Tune] Pin Transitive Dependencies (#13358) 2021-01-13 19:10:21 -08:00
Amog Kamsetty
43f70faa25
[Tune] Pin Tune Dependencies (#13027)
Co-authored-by: Ian <ian.rodney@gmail.com>
2021-01-07 14:03:06 -08:00
Ian Rodney
cf30630d2e
[docker] Use legacy resolver (#12741) 2020-12-10 01:12:46 -08:00
Ian Rodney
a9da4f3201
[docker] Make Ray-ml more compatible (#12574) 2020-12-09 16:03:39 -08:00
Ian Rodney
f5fe3794c8
[Docker] Uninstall Typing (#12500) 2020-11-30 14:12:57 -08:00
Ian Rodney
8f2b447ba4
[docker pipeline] Base-Deps, Dataclasses & Releases (#12119) 2020-11-18 14:34:04 -06:00
Lee moon soo
9920933e31
[docker] Support non-root container (#11407) 2020-11-12 08:41:50 -08:00
Ian Rodney
7a0184e081
[docker] Push to DockerHub in CI (#11442) 2020-10-23 12:02:15 -07:00
Ian Rodney
f37e967ada
[Docs] READmes for DockerHub (#11350) 2020-10-16 15:46:13 -07:00
Ian Rodney
6426fb3fff
[CI] Fix-Up Docker Build (Use Python) (#11139) 2020-10-12 14:22:51 -07:00
Ian Rodney
4324dd5929
[docker] Refactor "autoscaler" image into "-autoscaler" tag and "ray-ml" image. (#10351) 2020-09-02 13:03:35 -07:00