Edit

Run experiments

To access a list of training and test runs and explore their details, select AutoML > Experiments. Control Tower updates the list every time you start a training or a test.

Trainings

The Trainings section contains the following information on training runs:

Parameter	Description
ID	Model identifier specified in the course of its training. Training IDs correspond to model IDs. Once trainings are completed successufully, the IDs become links to the details of associated models.
Code	Identifier of the associated pipeline specified in its source code description.
Version	Version of the associated pipeline specified in its source code description.
Type	Problem type the model can solve (for instance, information extraction).
ML SDK version	ML SDK version used to build the associated pipeline
Started	Date and time of starting the training.
Finished	Date and time of completing the training.
Status	Current training status: Completed (green), Stopped (grey), Failed (error icon).

note

To update statuses during a session, switch to another page or refresh the Experiments page.

To view the data that interests you most, you can sort it by clicking the selected column.

Start training

To train a new model, in the top-right corner of the Experiments page, click the plus (+) button to open the menu and select Train new.

For the detailed flow on model training, refer to Manage models | Create model.

Filter and search

You can filter the list of displayed trainings based on any or all the four criteria:

• Code: Associated model code
• Type: Model type
• ML SDK version: ML SDK version used for building the model
• Status: Current training status

To apply the criteria, click one or more filter buttons at the top of the training list and select required items or find them using the search box. Filters turn blue as they become active, and the items in the training list change accordingly.

To turn off an active filter, click it again. To reset them all, click Clear all.

You can also search the list for a particular training by a model ID or a model code using the Search by ID and Code box in the top-right corner.

View error log

To view a log in case of an error, click the error indication.

A dialog with the error details appears.

To close the log, click the Close (cross) button or the Got it button.

Tests

The Tests section contains the following information on test runs:

Parameter	Description
Test ID	Test run identifier specified in the course of its running. Once tests are completed successufully, the IDs become links to the details of associated test runs.
Model ID	Identifier of the associated model.
Type	Problem type the model can solve (for instance, information extraction).
Started	Date and time of starting the test.
Finished	Date and time of completing the test.
Status	Current training status: Completed (green), Stopped (grey), Failed (error icon).

note

To update statuses during a session, switch to another page or refresh the Experiments page.

To view the data that interests you most, you can sort it by clicking the selected column.

Start test

You can apply your existing models to a test set to get metrics without coding and complex Business Process development.

1. In the top-right corner of the Experiments page, click the plus (+) button to open the menu and select Test new.

   

2. In the Model ID drop-down list, select the model by ID.

   

   If you create a test from the Models > the Actions menu > Test, the model ID is already selected.

   As soon as you specify the model ID, the pipeline and version are read from the model configuration and inserted automatically.

3. In the Test ID field, specify a unique name.

   

4. In the Test set list, select a test set.

5. Click Test.

   

   In AutoML > Experiments > Tests, find the new model test run: its status should be In progress.

   

6. Wait till the status changes to Completed. Then, click the test ID and review the resulting test data.

View test details

To access details of test runs that completed successfully, go to AutoML > Experiments > Tests and click a needed test ID.

The test run details are available in a separate window:

Here, you can explore the following test run data:

• Created: date and time when the test was created
• Model ID: link to the associated model
• Type: model type
• Test set: link to the test set used while testing
• Average for all label metrics (in %):
  • Precision
  • Recall
  • Accuracy
• Detailed statistics per label:
  • Label name
  • Label type
  • Label value
  • Metrics per label:
    • TP: number of true positives
    • FP: number of false positives
    • FN: number of false negatives
    • TN: number of true negatives
    • P: precision
    • R: recall
    • F1: f1-score
    • ER: error rate = FP/(FP + TN)
    • Accuracy

Filter and search

You can filter the list of displayed tests based on any or all the three criteria:

• Model: Associated model ID
• Type: Model type
• Status: Current test status

To apply the criteria, click one or more filter buttons at the top of the test list and select required items or find them using the search box. Filters turn blue as they become active, and the items in the test list change accordingly.

To turn off an active filter, click it again. To reset them all, click Clear all.

You can also search the list for a particular test by a test ID or a model ID using the Search by ID and Code box in the top-right corner.

View error log

To view a log in case of an error, click the error indication.

A dialog with the error details appears.

To close the log, click the Close (cross) button or the Got it button.