---
title: Make a one-time batch prediction
description: Make a batch prediction for a deployed model with a dataset of any size. Learn about additional prediction options for time series deployments.

---

# Make a one-time batch prediction {: #make-a-one-time-batch-prediction }

Use the **Deployments > Make Predictions** tab to efficiently score datasets with a deployed model by making batch predictions.

!!! note
    To make predictions with a model before deployment, select the model from the **Leaderboard** and navigate to [**Predict > Make Predictions**](predict).

Batch predictions are a method of making predictions with large datasets, in which you pass input data and get predictions for each row. DataRobot writes these predictions to output files. You can also:

* Schedule [Batch Prediction Jobs](batch-pred-jobs) by specifying the prediction data source and destination and determining when DataRobot runs the predictions.

* Make predictions with the [Batch Prediction API](batch-prediction-api/index).

##  Select a prediction source {: #select-a-prediction-source }

To make batch predictions with a deployed model, navigate to the deployment's **Predictions > Make Predictions** tab and upload a prediction source:

* Click and drag a file into the **Prediction source** group box.

* Click **Choose file** to upload a **Local file** or a dataset stored in the **AI Catalog**.

![](images/batch-2.png)

!!! note
    When uploading a prediction dataset, it is automatically stored in the **AI Catalog** once the upload is complete. Be sure not to navigate away from the page during the upload, or the dataset will not be stored in the catalog. If the dataset is still processing after the upload, that means DataRobot is [running EDA](eda-explained) on the dataset before it becomes available for use.

## Make predictions with a deployment {: #make-predictions-with-a-deployment }

This section explains how to use the **Make Predictions** tab to make batch predictions for standard deployments and time series deployments.

![](images/make-batch-predictions-ui.png)

| | Field name |  Description  |
|-|------------|---------------|
| ![](images/icon-1.png) | Prediction source | [Select a prediction source](#select-a-prediction-source) by uploading a local file or importing a dataset from the AI Catalog. |
| ![](images/icon-2.png) | Time series options | Specify and configure a [time series prediction method](#set-time-series-options).  |
| ![](images/icon-3.png) | Prediction options | [Configure the prediction options](#set-prediction-options). |
| ![](images/icon-4.png) | Compute and download predictions | [Score the data and download the predictions](#compute-and-download-predictions). |
| ![](images/icon-5.png) | Your recent predictions | View your recent batch predictions and download the results. These predictions are available for download for 48 hours.  |

## Set time series options {: #set-time-series-options }

{% include 'includes/batch-pred-ts-scoring-data-requirements.md' %}

{% include 'includes/batch-pred-ts-options-include.md' %}

## Set prediction options {: #set-prediction-options}

Once the file is uploaded, configure the **Prediction options**. Optionally, you can click **Show advanced options** to configure additional options.

{% include 'includes/prediction-options-include.md' %}

## Compute and download predictions {: #compute-and-download-predictions }

Once configured, click **Compute and download predictions** to start scoring the data.

![](images/batch-4.png)

When scoring completes, click **Download Predictions** to download a predictions file.

If the prediction job fails, click **View logs** to view and optionally copy the run details.  

Predictions are available for download on the **Predictions > Make Predictions** page for the next 48 hours. You can also view and download predictions and logs on the [**Deployments > Prediction Jobs** tab](batch-pred-jobs#manage-prediction-jobs).

!!! tip "Cancel a batch prediction job"
    Click the orange **X** while the job is running to cancel it. Once canceled, you can click the arrow to view the logs for the job.

    ![](images/batch-5.png)
