Add a CLI command to initialize kedro ipython with specified dataset_name as data variable

[yury-fedotov]

Situation

One operation I do very frequently is:

kedro ipython

...waiting for it to load. Then...

data = catalog.load("dataset_name")

And explore it somehow.

Idea

If that's a common scenario for many, maybe a CLI command like that may be in demand:

kedro ipython --fetch dataset_name

And it does the following:

1. Initializes kedro ipython
2. Performs data = catalog.load(dataset_name) automatically
3. You end up in ipython session just like after kedro ipython, with 1 difference: you have the data variable where a dataset you requested is available.

Alternatives

Instead of fetch, it can be provide.

[datajoely]

could you elaborate on what this would help you achieve?

[yury-fedotov]

could you elaborate on what this would help you achieve?

Just save ~ 10 mins per day and improve developer experience slightly.

Why I do this a lot of times per day:

kedro ipython
data = catalog.load("dataset_name")

Is to then do things like:

data.info()
data["Column"].describe()
(data["Column"] > 10).mean()
...
# And other things I need to look at to explore data, depending on what I'm modifying now.

So what my suggestion assumed is like a CLI alias that allows to do one command:

kedro ipython --fetch dataset_name

Instead of two

kedro ipython
data = catalog.load("dataset_name")