You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
From feedback from one of our collaborators (@nvander), it seems Karma does not produce error messaging in offline mode when a PyTransform fails, specifically when a particular package does not exist. This issue occurs because we are deploying Karma into a Docker environment and due to a bug one of our packages was not included in the environment.
Steps to reproduce the behavior:
Add a Python package into the Karma python folder
Load a dataset on WebKarma
Create a model on that data using the previously mentioned package.
Publish the model
Remove the package from the folder.
Apply the model to the dataset via offline mode
Karma will apply the model without errors, but will generate empty and erroneous RDF due to the missing package.
Expected behavior
Karma should return an error message when it encounters an issue applying a PyTransform. It should still proceed with applying the model, but the error message would be useful for users to identify problems in their environment or pipeline.
The text was updated successfully, but these errors were encountered:
From feedback from one of our collaborators (@nvander), it seems Karma does not produce error messaging in offline mode when a PyTransform fails, specifically when a particular package does not exist. This issue occurs because we are deploying Karma into a Docker environment and due to a bug one of our packages was not included in the environment.
Steps to reproduce the behavior:
Expected behavior
Karma should return an error message when it encounters an issue applying a PyTransform. It should still proceed with applying the model, but the error message would be useful for users to identify problems in their environment or pipeline.
The text was updated successfully, but these errors were encountered: