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
@sirspock, this will add new dependencies in the model catalog and will need to be discussed in detail. This new functionality implies the execution of data transformation and postprocessing steps for a model.
Assuming this information is available in the knowledge base, DAME will be able to invoke the transformation pipeline (dockerized) to produce inputs to be used by the model. Initially this will be done in 2 steps: the user activates the transformation giving the original input; and then executes the model with DAME.
Test the pull request Minh
Add Data Transformation resource to the Model Catalog issue
I do not have additional information, and I think you are right. We should engage with them soon or this feature will not be ready.
What I know is that they will provide a container with the transformations.
@sirspock, this will add new dependencies in the model catalog and will need to be discussed in detail. This new functionality implies the execution of data transformation and postprocessing steps for a model.
Assuming this information is available in the knowledge base, DAME will be able to invoke the transformation pipeline (dockerized) to produce inputs to be used by the model. Initially this will be done in 2 steps: the user activates the transformation giving the original input; and then executes the model with DAME.
Search dataset-id using the DataCatalog python clientThe text was updated successfully, but these errors were encountered: