search

mljar / mljar-supervised

Python package for AutoML on Tabular Data with Feature Engineering, Hyper-Parameters Tuning, Explanations and Automatic Documentation
https://mljar.com
MIT License
3k stars 401 forks source link

Change the distribution of discrete features in EDA #197

Closed pplonski closed 3 years ago

pplonski commented 3 years ago

Right now discrete features are plotted in the same way as categoricals in EDA plots. Please change that, they should be plotted the same way as continuous features.

And please remove the last dot in each feature description (near the picture).

pplonski commented 3 years ago

Could you take a look at this issue @shahules786 ?

pplonski commented 3 years ago

Example

image

the distribution of feature age looks strange and you can see the last 'dot' (should be removed).

pplonski commented 3 years ago

I'm working on tutorial how to integrate PostgreSQL with AutoML, here is the link for EDA results https://github.com/mljar/integrations/tree/master/PostgreSQL_AutoML/Response_Classifier/EDA would be nice to have results with this fix.