Predictions¶
After you deploy a model to production in the DataRobot Console, you can make predictions closely managed and monitored by DataRobot. The predictions you make through a deployed model provide the data displayed on the deployment's service health, data drift, and accuracy tabs, in addition to any other configured functionality.
| Topic | Description |
|---|---|
| Make predictions | Make predictions with large datasets, providing input data and receiving predictions for each row in the output data. |
| Portable predictions | Download and configure the Portable Prediction Server (PPS) or Scoring Code to make portable predictions. |
| Prediction API | Adapt downloadable DataRobot Python code to submit a CSV or JSON file for scoring and integrate it into a production application via the Prediction API. |
| Monitoring | Access monitoring snippets for agent-monitored external models deployed in Console. |
| Prediction intervals | For time series deployments, enable and configure prediction intervals returned alongside the prediction response of deployed models. |
| Prediction jobs | View and manage prediction job definitions for a deployment. |
Prediction file size limits¶
Self-Managed AI Platform limits
Prediction file size limits vary for Self-Managed AI Platform installations and limits are configurable.
| Prediction method | Details | File size limit |
|---|---|---|
| Leaderboard predictions | To make predictions on a non-deployed model using the UI, expand the model on the Leaderboard and select Predict > Make Predictions. Upload predictions from a local file, URL, data source, or the AI Catalog. You can also upload predictions using the modeling predictions API, also called the "V2 predictions API." Use this API to test predictions using your modeling workers on small datasets. Predictions can be limited to 100 requests per user, per hour, depending on your DataRobot package. | 1GB |
| Batch predictions (UI) | To make batch predictions using the UI, deploy a model and navigate to the deployment's Make Predictions tab (requires MLOps). | 5GB |
| Batch predictions (API) | The Batch Prediction API is optimized for high-throughput and contains production grade connectivity options that allow you to not only push data through the API, but also connect to the AI catalog, cloud storage, databases, or data warehouses (requires MLOps). | Unlimited |
| Prediction API (real-time) Dedicated Prediction Environment (DPE) |
To make real-time predictions on a deployed model, use the Prediction API. | 50MB |
| Prediction API (real-time) Serverless Prediction Environment |
To make real-time predictions on a deployed model, use the Prediction API. | 10MB |
| Prediction monitoring | While the Batch Prediction API isn't limited to a specific file size, prediction monitoring is still subject to an hourly rate limit. | 100MB / hour |