Feature Request: No ingest functionality to put multiple records of a data frame in feature group.
- Dominant language
- Python
- Stars
- 24
- Forks
- 25
- PR merge metrics
- No merged PRs in 30d
Description
In new SDK there is no `ingest` method to import a whole dataframe in a go creating difficulties in importing multiple records simultaneously.
Previous SDK had below code to inject a dataframe:
`customers_feature_group.ingest(data_frame=customer_data, max_workers=3, wait=True)`
sagemaker_core does not have this method/functionality. Without this, it will be difficult to import all records in feature store.
Old [sagemaker SDK document](https://sagemaker.readthedocs.io/en/stable/amazon_sagemaker_featurestore.html) containing ingest method.
If this functionality is not present in new SDK then how to mitigate this?
Contributor guide
Research direction
Review the new SDK's feature-group API in sagemaker-core and compare it with the previous SDK documentation linked in the issue. Determine how dataframe ingestion, max_workers, and wait are expected to work, and document or implement the agreed path for importing multiple records into a feature group.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100