Note
Access to this page requires authorization. You can try signing in or changing directories.
Access to this page requires authorization. You can try changing directories.
While training the category classification model, you might come across the messages in this article that AI Builder might report. Messages are either errors or warnings. Each is represented by an icon.
| Message | Icon |
|---|---|
| Error |  |
| Warning |  |
When an error occurs, you can't continue until you resolve it. If the system is unable to correct a problem, it will show you an error.
Warnings are messages reported as informational. They don't stop you from proceeding. They warn you of possible performance issues when training the model.
Error: InvalidTrainingInput
Cause
Following are the possible causes for receiving this error:
You've supplied fewer than 10 distinct training records per tag in your table.
You've supplied fewer than two tags, where each has 10 or more distinct training records in your table.
For each of your tags, you've supplied fewer than 10 distinct training records that don't contain the tag.
Resolution
Add a minimum of 10 distinct training records for each of the tags to be identified. Follow the guidelines in Before you build a category classification model to do the data preparation.
Error: No tag separator recommended
In the training step, you're prompted to choose No separator as the tag separator.
Cause
This error will occur if the tag separator used a mix of more than one separator.
Resolution
If you know you have data tagged with multiple tags, recheck the tag separator for each. You must use a single tag separator across all data rows.
Warning: Missing tags for some records
For the new records being created, you find that tags are missing for some records.
Cause
This will happen if you didn't provide a minimum of 10 sample text records for the tag while training the model.
Resolution
Add more sample text for the tag with missing data and retrain the model.