aws / aws/sagemaker-core

Feature Request: No ingest functionality to put multiple records of a data frame in feature group.

Open
#176 3 comments 0 reactions 0 assignees View on GitHub
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

Open the contributing 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

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.