Dataset¶
Predict allows you to create multiple datasets under each Predict asset. Each dataset is created with specific pattern, profile, features, normal or anomalous behaviors from the selected historic data records in the Predict asset of specfic time range, which can be used to create the engine of Prediction or Anomaly Detection.
Note
- "Predict asset": In this documentation, a "Predict asset" refers to any of your Insights Hub asset that you have connected for analytics (Predictions, Anomaly Detections). This term is used because the application organizes all analytics based on the specific asset being monitored.
- "engine": In this documentation, the term "engine" (e.g., "engine of prediction", "prediction engine", "anomaly detection engine") is used for domain experts from shopfloor. It is similar to what data scientists call a "model". This term is used because it better describes a powerful system that works continuously to analyze your assets and generate insights.
To create a new dataset, click "Add dataset" button on the Asset Based Home Page.

Creating a Dataset¶
To create a dataset, proceed with the following steps:
-
Open "Predict" by clicking on below icon on Insights Hub launchpad.

-
Click "Add asset" in the asset based home page to add a new Predict asset to the list.

Alternatively, you can also select an asset from the list in left and click "Add dataset" card in right section.

-
Enter the name and description.

-
Select the preferred date range.
-
Click "Add more data sources". In the "Add data source" page, select the Asset and Variables for the preferred dataset. Click "Add data source".
-
Click "Save".
A dataset to create engines of Predictions and Anomaly Detections is created.
Editing or Deleting a Dataset¶
To edit an existing dataset, proceed as follows:
- In the "Dataset" screen, select the existing dataset.
- Click
and click "Edit". - In the Edit Dataset screen, update the required fields.
- Click "Save".
To delete an existing Dataset, select the dataset, click
and click "Delete". In the confirmation Dialog, click "Delete".