-
Move projects between folders option
The move projects between folders option does not show in the Dataiku home page and only shows when you visit a project while navigating a project folder or when you visit /project-list/. This is a silly UX inconsistency which is compounded by the fact that you can't really see the project folder full path in the Dataiku…
-
remove the depenency between DSS pandas and python recipe pandas
Dataiku demands pandas>=1.1,<1.2. Bokeh 3.* demands pandas>=1.2 geopandas 0.14.* demands pandas>=1.4 So, they can't be used within DataIku. I think there are other examples. I imagine there are several solutions, one which seem simple is to create a fork of pandas, e.g. dsspandas, an exact copy of pandas 1.1 and replace…
-
Rename tables in Snowflake after renaming a dataset
I recently tried renaming and running a dataset in Dataiku and noticed this did not rename the table in Snowflake. I believe this can lead to confusing situations in case you rename a dataset and there is a mismatch between Snowflake table and Dataiku dataset.
-
Changing block flows
It should be more intuitive to change the blocks in the flow. Especially with code blocks where you actually need to change the code if the input changes. There should be like identifier of input blocks (like DataInput_1), that you use in the code instead of using the name of the previous stage (that might change).
-
Permit execution of pre-write and post-write statements in settings/advanced Datasets definition
I have a dataset which takes 20 minutes to be generated. To improve reading performance (it is used in a webapp), I've tried several indexes definitions. It's boring to await the full execution of the parent recipe to generate the new indexes. (and exasperating when you misspelled the post-write statement) A simple…
-
Convert Boolean Variables to 0/1
When I bring Boolean variables to my model, it treats them as Categorical. The default option is to dummy encode which makes a column of var_false and var_true. It should convert the boolean in the column to 0 and 1, to leave it as one column, not make a second redundant column.
-
Document the start_tool flow options
A user can trigger a flow consistency check using the following API code: import dataikuclient = dataiku.api_client() project_handle = client.get_project("some project key")flow_handle = project_handle.get_flow()tool = flow_handle.start_tool(type='CHECK_CONSISTENCY')options = { "recheckAll": False, "datasets":…
-
Year not displayed on Code Env Last Modified column
The Year is not displayed on the Code Env Last Modified column. As a result the Last Modified is misleading since it says the day and month but not the year. As a sample in the attached screen shot you see "Wednesday, 29 March" however this is not for the current year and you have to look at the logs to find out what year…
-
Analyze - more value and actions
In a Project, under Explore, then Analyze, Numerical, there is the little option at the bottom to expand "more value and actions". This cannot be undone / reversed unless the window is closed and you start again.
-
SQL Notebooks: Allow for Tables/Queries left pane to be resizeable
In SQL Notebooks we would want the Tables/Queries left pane to be resizeable. This is because our table names are preffix with the environment name and then the project ID which means that there are only a few characters left visible for the table name, sometimes not even that (see sample below). We would like to be able…