Small typo

[Mirzon]

The note at the bottom of the page asked me to report any kind of mistakes encountered in this course, so here is a small one :-).

I believe it should be “our” instead of “out”.

[ThomasLoock]

Hi mirzon, i assume it to be correct. Look here: Separate Out | Definition of Separate Out by Merriam-Webster

[Mirzon]

Oh, right, I parsed the sentence wrong, I assumed “out” was grouped with “input” instead of “separate”, thanks for the explanation!

[Marc_In_Singapore]

Exercise M1.03. Instructions for the last cell. It should be the parameter stratified of the Dummy estimator rather than strategy.

3.3.6. Dummy estimators

[Mirzon]

The name of the parameter is strategy, stratified is just one of the possible values it can take:
https://scikit-learn.org/stable/modules/generated/sklearn.dummy.DummyClassifier.html#sklearn.dummy.DummyClassifier

And in our case we are more interested by the constant strategy than the stratified strategy.

[Marc_In_Singapore]

Got it. Read too fast. Thanks

[glemaitre58]

Fixed in FIX typo · INRIA/scikit-learn-mooc@ba622d3 · GitHub

[blaiseli]

I’ve spotted an OrdinaleEncoder in the first notebook of the “Handling categorical data” section of Module 1 (“Encoding of categorical variables”).
It should be OrdinalEncoder, without the extra “e”.

[Marc_In_Singapore]

Exercise M4.02. Typo in the tip; should be “to get non-deterministic results”, maybe.

[glemaitre58]

Uhm. I don’t think so. We fix the seed of the random number generator to control the randomness and thus make deterministic results between code run.

[Marc_In_Singapore]

Got it.

[lesteve]

@blaiseli thanks this one has been fixed by @glemaitre58. You can reset your notebook to the latest version following How to reset a notebook to its original version?.

[lesteve]

General comment: this is great if you can create new topics for new problems. This makes it a lot easier to take them into account for us !

For example this topic has been marked solved but since many different things were reported we are not quite sure what still needs to be done.

Also this topic has the category “Module 1. The Predictive Modeling Pipeline → Fitting a scikit-learn model on numerical data” but some posts were about some other parts of the MOOC.

Note: you can see the category just below the topic title:

I think only Small typo - #8 is left to tackle, I am going to move it to a new topic so that we remember to do it.

[reshama]

In the quiz explanation, there is a small typo
“to pass” should be “to pass”

[glemaitre58]

@reshama Could you provide the title of the quiz title and copy-paste the sentence such that I can find where it is. I assume that you had an auto-correction that make it difficult to note the pattern to search

[reshama]

Quiz M1.02

Question 3

This is in “Explanation” section:

By default, the estimators or the train scores are not returned. One needs topass return_estimator=True and return_train_score=True . We will present these options in an upcoming lesson. For more details, regarding cross_validate , you can refer to the documentation.

[glemaitre58]

Thank you @reshama . @lfarhi @MarieCollin this typo is only in FUN. Could you correct it?

[MarieCollin]

Thanks @glemaitre58 and @reshama, the typo is now fixed in FUN.