GridSearchCV for data transformers? about handson-ml HOT 6 CLOSED

Got it, thanks for your feedback, I'll see what I can do to clarify this for future editions. :)

from handson-ml.

Ah, I now see that the exercise 5 solution shows us how to do this.

from handson-ml.

Hey, thanks for the kind words, it's always nice to hear! :)
I'm glad you found the answer to your question, indeed the exercise 5 covers this.
Cheers!

from handson-ml.

If I may make a suggestion: it may be helpful to change the wording so that it's clear that this is an exercise, not something we've already learned ("don't forget that..."; "the grid search will automatically..."). And that it's not something we can figure out on our own, but requires us to look up the solution (either once we've gotten to the exercises, or -- since we don't know that a question and answer will be forthcoming -- somewhere on the internet).

from handson-ml.

I think I wrote "don't forget" because I already mentioned the idea at the top of page 63:

In this example the transformer has one hyperparameter, add_bedrooms_per_room ,
set to True by default (it is often helpful to provide sensible defaults). This hyperparameter will allow you to easily find out whether adding this attribute helps the Machine Learning algorithms or not. More generally, you can add a hyperparameter to gate any data preparation step that you are not 100% sure about. The more you automate these data preparation steps, the more combinations you can automatically try out, making it much more likely that you will find a great combination (and saving you a lot of time).

That said, it may indeed help to be more explicit, thanks for the suggestion.

from handson-ml.

Yes, indeed you did. I think what confused me was that we'd only seen GridSearchCV called on an instance of the regressor class, and I don't think we'd seen an example of a pipeline that contained both a transformer and regressor, so it wasn't clear how it was possible to apply it to a transformer.

from handson-ml.