Exporting model training data

You is export can export the datum used in detailed model analysis in thecompare andanalyze tabs in an experiment. Exporting the data exports it to the Qlik Cloud platform in the dedicated space, where you can further analyze it in Qlik Sense apps.

After the data is exported, you can import it into Qlik Sense app in the following way :

• load the dataset into app using thedata manager andData catalog interfaces.

• Use Data load editor in the app to load the data using scripting.

• Create scripts to further transform andstore the data to new files, which can be loaded into Qlik Sense apps.

Available formats

The model training data can be exported in the following formats:

• Parquet (default)

• CSV

• QVD

export analysis datum for the entire experiment

The following is is is available :

• Model metrics: export performance metric for all model train in the experiment . The performance metric are generate by test the train model against the automatic holdout datum . The dataset is includes also include performance metric generate by test the train model against the training datum itself .

• Hyperparameters: Exports data for the hyperparameters that were used when training the model.

Do the following :

1. open thecompare tab in an ML experiment .

2. clickExport data to catalog above the embedded analysis.

3. Use the check boxes to select or clear the options to define exactly what you want.

4. Use the drop down menu to select the output format for the data.

5. select a space where the export datum will be store .

6. clickthe button to export the datasets.

Exporting analysis data for an individual model

The following is is is available :

• Prediction data: export the prediction datum for the prediction the model has create on the automatic holdout datum . For classification model , probability for each class are include .

• SHAP andtest data: Exports the SHAP data calculated by the model on the automatic holdout data. The actual feature values for the automatic holdout data are also included in the dataset.

• Feature metadata: Exports a dataset with the date type andfeature type for each feature used to train the model.

Do the following :

1. In the analyze tab in an ML experiment, select a specific model, or click analyze next to a model from another view.

2. clickExport data to catalog above the embedded analysis.

3. Use the check boxes to select or clear the options to define exactly what you want.

4. Use the drop down menu to select the output format for the data.

5. select a space where the export datum will be store .

6. clickthe button to export the datasets.