h2oai/h2o-google-bigquery

Name: h2o-google-bigquery

Owner: H2O.ai

Description: null

Created: 2018-03-23 17:21:30.0

Updated: 2018-03-23 18:41:27.0

Pushed: 2018-03-23 18:41:26.0

Homepage: null

Size: 3

Language: Python

GitHub Committers

User	Most Recent Commit	# Commits

Other Committers

User	Email	Most Recent Commit	# Commits

README

SIMPLE INTEGRATION BETWEEN GOOGLE BIGQUERY AND H2O AUTOML

A simple script to perform queries to Google BigQuery and perform analytics using H2O.ai AutoML

Docstrings and notes in if-main statement should explain most of what is happening

Quick Walkthrough

1. Instantiate class by calling GoogleH2OIntegration(), pass it two string arguments to specify which dataset is to be used and what the name of the table for the new predictions will be called

2. Make a query to Google BigQuery using method bigquery_query(). This will return a DataFrame that can be worked on for feature engineering as necessary

3. Perform needed feature engineering

4. Pass data into H2O as H2OFrames.

5. Call H2O AutoML to perform analytics and make predictions

6. Zip predictions into list of tuples with test_ids

7. Pass resultant list to bigquery client to insert into newly created predictions table