Kedro is an open-sourced Python framework for creating maintainable and modular data science code.

Kedro

image.png

I've just been playing with `kedro&gt;=0.19.0` - what is creating these two files when I run?

I can't remember if it's the correct location of the file, <@U03SNB75DDE> do you know about this?

On the other hand, <@U0485D1780J> It's kedro-viz, since you probably select `all` when you run `kedro new`? By default it should be `None`

This is because we assume if you select `kedro-viz` you want to have it turns on automatically.  This files appear after your first run.

Also - might be good to have them in a `.viz` folder or similar?

`session_store.db` is the same old experiment tracking feature, so it will keep all runs history.

`stats.json` is something we added a few months ago in `kedro-viz`, it's refresh on each run. (the flowchart view of viz)

Is there any convenient way to open and look at `session_store`.db in python?

It's a sqlite database, so something like this should work. <https://datacarpentry.org/python-ecology-lesson/instructor/09-working-with-sql.html>

There isn't too much information, it's store the information that you see in the experiment tracking tab

<https://demo.kedro.org/experiment-tracking>

My prefer option to peek at the db is this tool. <https://github.com/simonw/datasette>

We don’t have any logic in our cookiecutter hooks to deal with these files so they are in their default location. As <@U0485D1780J> suggested might be nice to have them in a separate directory.