Details
-
Sub-task
-
Status: Resolved
-
Major
-
Resolution: Incomplete
-
2.3.0
-
None
Description
In pyspark:
We add a parameter indicating whether to persist models to disk during training (default = off). This will use ML persistence to dump models to a directory so they are available later but do not consume memory.
Note: when persisting the model list, use indices as the sub-model path
Attachments
Issue Links
- is blocked by
-
SPARK-21088 CrossValidator, TrainValidationSplit should collect all models when fitting: Python API
- Resolved
- is required by
-
SPARK-23109 ML 2.3 QA: API: Python API coverage
- Resolved