more info about "highly correlated" #960
Unanswered
anaflorido
asked this question in
Q&A
Replies: 1 comment
-
Starting of with a short explanation of how the alerts are generated. The thresholds for this warning are set per correlation, and their defaults can be found here (0.9). Sensible values for the threshold may differ per dataset. In practice, the dendrogram under 'missing values' is provides an intuitive view for similar variables in terms of missing values. (Hierarchical clustering of the correlation matrices could be an interesting feature!) As to your second question, could you please clarify what the question is? |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hi, im using pandas profiling for data exploration (within google colab notebook with "!pip install https://github.com/pandas-profiling/pandas-profiling/archive/master.zip"), and the thing is, i cant find any details about "highly correlated" on the library documentation,
i would like to know more info about this thresholds that pandas profiling is using when finds a "highly" or other correlations (like > 0.7? less than 0.3? etc)
and when i get a full profile, and i cant see the values list of corr either: do you know if there's a flag for this?
thanks!
Beta Was this translation helpful? Give feedback.
All reactions