Skip to content
#

feature-engineering

Here are 1,044 public repositories matching this topic...

nni
Agent007
Agent007 commented Jun 11, 2021

Expected Behavior

1-hot encoded features should be materializeable to the online store without error.

Current Behavior

Version 0.10 apparently introduced a regression where 1-hot encoded feature materialization throws errors.

Steps to reproduce

from datetime import datetime, timedelta
from feast import Entity, Feature, FeatureStore, FeatureView, FileSource, ValueType
imp
evalml
dsherry
dsherry commented Jun 17, 2021

The "Unit tests, linux, min dependencies" workflow currently runs a "shellcheck" job. The "Unit tests, linux, latest dependencies" and "Unit tests, windows" workflows do not.

If that check is not necessary, let's delete it. If it is necessary, let's add it in all the unit test workflows.

I think this came in when @gsheni recently merged the min deps CI.

Feature engineering is the process of using domain knowledge to extract features from raw data via data mining techniques. These features can be used to improve the performance of machine learning algorithms. Feature engineering can be considered as applied machine learning itself.

  • Updated Nov 29, 2020
  • Jupyter Notebook
Hyperactive

Improve this page

Add a description, image, and links to the feature-engineering topic page so that developers can more easily learn about it.

Curate this topic

Add this topic to your repo

To associate your repository with the feature-engineering topic, visit your repo's landing page and select "manage topics."

Learn more