Dataset and ML models dashboard

piyushk

Can we get the overall dataset count and ML model usages in a project for different projects such that dashboard can be made for them?

We can see these information in Administration -> Monitoring -> Summary. We are looking to get similar information in the form of dashboard in a project.

Further, can we get detailed information for ML models used in different projects in a single project?

Andrey

Hi Piyush,

I'm going to copy here my answer from the support ticket for the good of the community.

There's a Public API that you could use. The easiest way is to communicate with it is through the Dataiku Python client. In order to obtain the information you were looking for you could use: 

import dataiku, dataikuapi

c = dataiku.api_client()
datasets_per_project = {pk: dataikuapi.dss.project.DSSProject(c,pk).list_datasets() for pk in c.list_project_keys()}
saved_models_per_project = {pk: dataikuapi.dss.project.DSSProject(c,pk).list_saved_models() for pk in c.list_project_keys()}
​

for more details about the Python client please refer to this page:

https://doc.dataiku.com/dss/latest/python-api/rest-api-client/index.html

So if you want to display this information on a dashboard you could create a notebook using a snippet above + some markdown for styling:

Next, create a notebook tile on a dashboard, it'll look like this:

 

and finally, in order to refresh the statistics regularly, you may want to create a time-based scenario with a "Create notebook export" step.

Best Regards