Antoni Baum
ea94cda1f3
[AIR] Replace train. with session. ( #26303 )
...
This PR replaces legacy API calls to `train.` with AIR `session.` in Train code, examples and docs.
Depends on https://github.com/ray-project/ray/pull/25735
2022-07-07 16:29:04 -07:00
Simon Mo
88a219c7f2
Revert "Revert "[AIR][Serve] Rename ModelWrapperDeployment -> PredictorDeployment"" ( #26231 )
2022-07-05 13:26:49 -07:00
Stephanie Wang
c9be251b7a
Revert "[AIR][Serve] Rename ModelWrapperDeployment -> PredictorDeployment ( #25962 )" ( #26176 )
...
This reverts commit 68692b3464
2022-06-28 17:07:07 -07:00
Simon Mo
68692b3464
[AIR][Serve] Rename ModelWrapperDeployment -> PredictorDeployment ( #25962 )
2022-06-28 10:26:10 -07:00
Antoni Baum
91dd360f9d
[AIR/train] Move predictors to ray.train ( #25769 )
2022-06-15 17:02:15 -07:00
Antoni Baum
5e9a8eb5f6
[AIR/data] Move preprocessors to ray.data ( #25599 )
...
Moves ray.air.Preprocessor and ray.air.preprocessors to ray.data to converge on the agreed upon package structure discussed internally.
2022-06-13 12:57:59 -07:00
Amog Kamsetty
1316a2d05e
[AIR/Train] Move ray.air.train to ray.train ( #25570 )
2022-06-08 21:34:18 -07:00
Kai Fricke
4b9a89ad90
[air] Move python/ray/ml to python/ray/air ( #25449 )
...
The package "ml" should be renamed to "air".
Main question: Keep a `ml.py` with `from ray.air import *` for some level of backwards compatibility?
I'd go for no to force people to use the new structure.
2022-06-03 21:53:44 +01:00
Amog Kamsetty
e8440cf52b
[AIR] Incremental Learning Example ( #24420 )
...
Example for domain incremental learning on Permuted MNIST Dataset with naive strategy
2022-05-26 12:28:28 -07:00
