-
Maintain case of SQL table name when creating SQL datasets
Currently, when a SQL dataset is created, the name of the associated SQL table is set to PROJECTKEY_tablename regardless of the case of the SQL dataset name. It would be great if either the case of the dataset name was maintained in the SQL table name (so dataset ABC would result in a SQL table name of PROJECTKEY_ABC…
-
Support image segmentation in labelling tasks and Visual ML
I have a couple of use cases where I need to train image (instance) segmentation models (as opposed to predicting bounding boxes in object detection). I'd love for the ML labelling to support image segmentation approaches. For example using SAM (Segment Anything Model) to pre-segment images which can then be annotated by…
-
Support fot 2way partial dependence plots
I'd love to see support for 2way partial dependence plots in mode summary reports to get insights into the interaction of 2 features on their model impact. This would give some deeper insight into feature behavior in the model at hand. See here under 4.1.1 for the sklearn implementation 4.1. Partial Dependence and…
-
Select which test queries to run in API Designer
Often times when testing and adjusting test queries, I only want a single query to be executed. It would be helpful to have selection boxes next to the test queries in API Designer for which queries to (re-)run when using the Run testqueries command.
-
Deployment hooks should be able to have an impact on the status of project deployments.
Hello, We have found that the post-deployment hooks have no impact on the project status on the deployer at this time.We have a confirmation from dataiku support. In other words, if you deploy a dataiku project in an automation node infrastructure, if you have configured post-deployment operations by creating hooks in the…
-
Backup execution for Kubernetes
Ran into a Microsoft issue with connectivity in Azure, which forced me to manually move over to local execution. Idea: When an execution is set to run in a pod but fails, reattempt 'n' times, then upon failure, default to local execution to run. Error handling by default from local execution. Option to send admin an email…
-
Idea: Include Associated Objects When Duplicating Dataset / Flows
Hello When duplicating parts of a flow in Dataiku, the associated datasets are duplicated, but the developed charts linked to those datasets are not included. This means that users have to manually recreate or copy these charts, which can be time-consuming and prone to errors. Benefits Including the duplicate feature for…
-
Improve Tasks in Project Todo List
I would really like to use the todo list accessible in a Dataiku DSS project home page to communicate with my team. However it is not always easy to use it. Here are a couple of improvements suggestions: Allow tasks to be moved / reordered The text editing area has a fixed maximum height, which makes it difficult to edit,…
-
Invalid Scenario step logic condition should cause scenario failure
I have noted a very dangerous behavior in the latest v12 release although I believe this has been in DSS for a long while. DSS will not cause a scenario failure or even a warning if you have an invalid Scenario step logic. For instance I created a scenario step and set two variables: {"var1": 123, "var2": 456} Then I…
-
Being able to set User Settings via Python API
There are now more user settings than ever in the user's profile page (DSS/profile/). In v12.3.2 there are now 10 different email notification settings. We would like to be able to customise these to our preferred defaults via the Dataiku Python API. Currently the Dataiku Python API does not support this. Thanks