System status: Everything is great. Last updated May 24, 2022 at 10:30am MT. |
If you're experiencing a problem with Qlik AutoML, this is your first stop to determine if we're already aware of the situation. Whenever possible, we provide suggested workarounds while we squash bugs and resolve issues. This article will be updated as we provide fixes through platform updates, as well as when new issues are identified.
If you have any questions after you've reviewed the material, contact Qlik AutoML Support or your friendly neighborhood Customer Success Manager for assistance.
Clear Your Browser Cache (do this first!)
- If you are experiencing an issue, especially if you encounter a generic "Something went wrong" error message, there may be stale content in your browser cache. To verify that your cache is not the problem, try refreshing your browser page. If that doesn't resolve the issue, clear your browser cache, quit and restart your browser, then log in to Qlik AutoML. As an additional step after you restart your browser, you can use an Incognito/Private browsing window temporarily to ensure that browser cache isn't causing a problem. [KRKN-1245]
Data Providers
All Data Providers
- When viewing or refreshing the list of available datasets, you may see one or more "ghost" datasets that have been deleted from your data provider and are no longer available for analysis or predictions. When you select one of these datasets, a yellow caution icon will be displayed on the right-hand side of the screen, along with a "Error requesting stats. sql: no rows in result set" hover message. Because the dataset is not actually available, you will be unable to proceed with using the dataset for analysis or predictions. This situation will be resolved in a future update, so that datasets that are no longer available will not be displayed in the dataset list. [KRKN-1410]
- To avoid unexpected errors, ensure that your datasets - both training and apply - do not contain carriage returns and other non-standard "new line" characters. [KRKN-1166]
- To avoid unexpected "internal error" messages, ensure that your datasets - both training and apply - do not contain non-English UTF-8 special characters, such as "ô" and "ü". [KRKN-1244]
Domo
- Domo datasets with a forward slash ("/") in the dataset name may result in the ingestion of two datasets, each named on either side of the slash. Avoid using a slash in the dataset name. [KRKN-753]
- If your dataset contains column names with spaces (e.g. "Annual Revenue") your dataset may fail to be processed correctly. Avoid spaces in column names. [KRKN-754]
Snowflake
- Special characters in Snowflake column names may prevent the dataset from being processed correctly. Avoid using < > [ ] { } characters in your column names in Snowflake tables. [KRKN-793]
Model Metrics
Correlations and Target Correlations cards
- If you use a dataset with more than 50 features and refine your Analysis to use less than 50 features, you may encounter a "Your model contains too many columns to display this chart. Refine your model and reduce the number of columns to 50 or less." message. To avoid this situation, create your initial analysis using a dataset with 50 or less features. [KRKN-1804]
Hyperparameter values for legacy models (pre-2020)
- If your binary, multiclass or regression Analysis was created before January 1, 2020, it is highly likely that the Model Metrics details will not display the hyperparameter values used in the Analysis. Instead, you will see a "Retrain for Hyperparameters" message in the Model Metrics screen. Prior to 2020, the hyperparameter values used for model training were not recorded, so there are no hyperparameter values to display. If you retrain your model, you will see the hyperparameter values used during retraining when you access the Model Metrics screen.
Before retraining
After retraining
Model Retraining
- If you encounter a situation where the Analysis "Retrain" button doesn't trigger a retrain, you can click the "Refine" button (next to the "Retrain" button) and then click the "Analyze" button in the upper right to retrain your Analysis. [BUGS-55]
Project Sharing
- Before accepting a Project share from another user (by clicking either the link or the button in the share invite email) you need to be logged in to the AutoML platform. If you are prompted to login when accepting the share, the share will not be successful; accepting the share when you're already logged in will result in a successful share. [BUGS-47]
- When sharing a Project with a user, the Project share email may not successfully be sent. If this occurs, you can contact Support for assistance in getting your Project manually shared with the desired user(s). [BUGS-52]