---
title: Batch scoring (video)
description: Watch a series of videos to learn how to set up scheduled batch scoring with AWS, Azure, BigQuery, and Snowflake.

---

# Batch Scoring (video) {: #batch-scoring-video }

Because recurring batch scoring to a data warehouse is a common way to consume model predictions, DataRobot supports executing prediction jobs on a repeatable schedule and writing the predictions to an external table. The write process can be set to update or overwrite; it can even create a new table for you if the table does not already exist. When configuring the batch job, you can select which of the features from the prediction data set to include and choose to include up to 10 [Prediction Explanations](pred-explain/index). These videos show the set-up process required to get batch scoring running on a schedule with some of our cloud and tech alliance [partners](https://www.datarobot.com/partners/){ target=_blank }.

* [AWS S3](#score-from-aws-s3)
* [Azure Cloud](#score-from-azure-cloud)
* [GCP BigQuery](#score-from-gcp-bigquery)
* [Snowflake](#score-from-snowflake)

<hr>


## Score from AWS S3 {: #score-from-aws-s3 }

This video shows how to set up a batch scoring job where the data source and the prediction destination is Amazon S3. Specifically, it pulls in data from Amazon S3, runs inference, and writes predictions back to S3 tables.

<div style="position:relative;padding-bottom:56.25%;">
 <iframe style="width:100%;height:100%;position:absolute;left:0px;top:0px;" title="Model lifecycle management" frameborder="0" width="100%" height="100%"
 allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" src="https://www.youtube.com/embed/DM9Hbls18Ro" allowfullscreen></iframe>
</div>
<br>

<hr>


## Score from Azure Cloud {: #score-from-azure-cloud }

This video shows how to set up a batch scoring job where the data source and the prediction destination is Azure SQL. Specifically, it pulls in data from Azure blob storage, runs inference, and writes back to Azure tables.

<div style="position:relative;padding-bottom:56.25%;">
 <iframe style="width:100%;height:100%;position:absolute;left:0px;top:0px;" title="Deployment workflow options" frameborder="0" width="100%" height="100%"
 allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" src="https://www.youtube.com/embed/y8b-NK33PN4" allowfullscreen></iframe>
</div>
<br>

<hr>


## Score from GCP BigQuery {: #score-from-gcp-bigquery }

This video shows how to set up a batch scoring job where the data source and the prediction destination is BigQuery. Specifically, it pulls in data for inference from GCP, makes predictions, and writes the predictions back to GCP BigQuery.

<div style="position:relative;padding-bottom:56.25%;">
 <iframe style="width:100%;height:100%;position:absolute;left:0px;top:0px;" title="Deployment workflow options" frameborder="0" width="100%" height="100%"
 allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" src="https://www.youtube.com/embed/PXfQcmdderU" allowfullscreen></iframe>
</div>
<br>

<hr>


## Score from Snowflake {: #score-from-snowflake }

This video shows how to set up a batch scoring job where the data source and the prediction destination is Snowflake. Specifically, it pulls in data from Snowflake for inference and writes the predictions back to Snowflake. See also [Snowflake prediction job examples](pred-job-examples-snowflake).

<div style="position:relative;padding-bottom:56.25%;">
 <iframe style="width:100%;height:100%;position:absolute;left:0px;top:0px;" title="Deployment workflow options" frameborder="0" width="100%" height="100%"
 allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" src="https://www.youtube.com/embed/EJeuKU2WsYI" allowfullscreen></iframe>
</div>
<br>

<hr>

## Read more {: #read-more }

* [Batch prediction methods](batch/index){ target=_blank }
* [Data connections](data-sources/index){ target=_blank }
* [Schedule recurring batch prediction jobs](batch-pred-jobs){ target=_blank }
* [Prediction intake options](intake-options){ target=_blank }
* [Code examples to make batch predictions](gcs-pred){ target=_blank }
* [Large batch scoring and object storage](sf-large-batch){ target=_blank }

Also, end-to-end workflow AI Accelerators:

* [AWS](ml-aws){ target=_blank }
* [Azure](ml-azure){ target=_blank }
* [BigQuery](ml-gcp){ target=_blank }
* [Snowflake](ml-snowflake){ target=_blank }
