How can i identify the main components? #156
-
|
I'musing the famd with the load beers dataset, but I have one doubt: how can I identify the main components? |
Beta Was this translation helpful? Give feedback.
Answered by
MaxHalford
Jun 22, 2023
Replies: 2 comments 1 reply
-
|
You can access the |
Beta Was this translation helpful? Give feedback.
0 replies
-
|
dear Max,
i'm familiar with the eigenvalues_summary, but it only shows the
eigenvalues, percent of variance (non and cumulative) of the desired number
of components; but what i'd like to know is: based on the data given, who's
the component 0? who's the component 1? And so on.
For instance, for the beers dataset, I'd like to see something like this:
component 1 - "alcohol by volume"
component 2 - "final gravity"
…--
Felipe Nunes Radtke
***@***.***
Em qui., 22 de jun. de 2023 às 04:12, Max Halford ***@***.***>
escreveu:
You can access the eigenvalues_summary property of your FAMD instance.
—
Reply to this email directly, view it on GitHub
<#156 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/BAX6FSWYOSDAZYQHIPETUHTXMPV7DANCNFSM6AAAAAAZPMRJSE>
.
You are receiving this because you authored the thread.Message ID:
***@***.***>
|
Beta Was this translation helpful? Give feedback.
1 reply
-
These aren't components, they're the columns in your original dataset. I think what you want is the contribution of each column to every component, i.e. the |
Beta Was this translation helpful? Give feedback.
Answer selected by
feliperad
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
These aren't components, they're the columns in your original dataset. I think what you want is the contribution of each column to every component, i.e. the
column_contributions_property.