[?] Delete missing values of a dataset?

[ttfreitas (Thadeu Freitas Filho)]

Hello, Mathematica Community
I have imported the following data as a DataSet into Mathematica

and I’m trying to run the following Linear Model

When I run the following code, I get all that missing data

But I’m unable to select the rows with missing data

Nothing shows up.
Can someone please help me to select those rows and delete them?
Because when I examine the original data, I can’t find the rows with missing data either so I am curious to see Mathematica’s output.
Thank you in advance,

Thad

[chrispakleza (Chris P)]

Probably, the missing data is tagged with additional “reason” (look at the docs for the Missing syntax).

Maybe this example will help:

ds = Dataset[{<|"a" -> 1, "b" -> 11|>, <|"a" -> 2, "b" -> Missing["somereason"]|>, <|"a" -> 3, "b" -> Missing[]|>}]

ds[Count[_Missing], "b"]

2

ds[Select[#b == Missing[] &]]

ds[Select[#b == Missing["somereason"] &]]

ds[Select[MatchQ[#b, _Missing] &]]

[ttfreitas (Thadeu Freitas Filho)]

Thanks, Chris P
And I think you are spot on because when I investigate further, I see missing values on the DataSet (house$size)

that are not missing in the original data (notice house$size and nbaths as well)

What’s going on?
Do you know how I can fix this?

Thanks,
Thad

[chrispakleza (Chris P)]

If I understand well, the problem seems to be when you import the original data into Mathematica. I just remark that the missing nbaths value is a real (1.5) whereas all the others are integers (1) …

Anyway, you should start a new question, telling precisely how you imported the data, what format the original data are, and give a minimal example.

[ttfreitas (Thadeu Freitas Filho)]

Thanks, Chris P
I imported the data from a csv type format (filename = train.csv) using

data = SemanticImport[ “C:\\Users\\Thadeu\\Documents\\Kaggle\\train.csv”]

I’ll start a new question.
Thanks