[Tutor] Explanation of one - hot encoding

[Manprit Singh]

Dear sir,

One hot encoding is basically a preprocessing step in machine learning ,
where a categorical feature is encoded as a numeric array of 0 and 1 . One
hot encoding is not considered good when a feature has a large number of
unique values in it .

Regards


On Sat, Nov 27, 2021 at 5:08 PM Tariq Khasiri <tariqkhasiri at gmail.com>
wrote:

> This following line is from a kernel I am trying to replicate from Spotify
> songs data. Could anyone kindly explain what this line means ?
>
> “ In terms of encoding our artists and our genres we see one-hot encoding
> is a bad idea here considering the cardinality of those features therefore
> we will dummy encode the genres and replace the artist names with numerical
> values closer to the model selection stage. ”
> _______________________________________________
> Tutor maillist  -  Tutor at python.org
> To unsubscribe or change subscription options:
> https://mail.python.org/mailman/listinfo/tutor
>