I am working on a data set that has 10 features, and my label ed output is the 'Weight of humans'.
I want to find, out of 10 features, which are he 2 or 3 features due to which the 'Weight' varies.. What are these 2-3 features on which the weight varies...
Please let me know, how i can find it out, using any ML technique or algorithm..
Tags:
Hi, I'm a new student in data science, I'm doing the process the following way.
from sklearn.feature_selection import RFE
from sklearn.linear_model import LogisticRegression
modelo = LogisticRegression()
rfe = RFE(modelo, 8)
fit = rfe.fit(X,Y)
print(X.columns)
print(fit.support_)
print(fit.ranking_)
melhores_colunas = X.columns[fit.support_]
Depends on what algorithm you are using. Assuming this is a classification problem and you are using RandomForestClassifier from sklearn, you can simply use it feature_importances_ method to look at the a sorted list of the features and determine which are more important.
Sklearn RandomForestClassifier: https://scikit-learn.org/stable/modules/generated/sklearn.ensemble....
Checkout this blog post- https://blog.datadive.net/selecting-good-features-part-iii-random-f...
Although it talks about RandomForestRegressor but it may help you understand this...
Thanks the above links are helpful, but here my case is a regression one, but here i don't want to predict anything, i want to know features that are impacting my label ed output.
Have you tried looking at a correlation in the variables? Maybe you could use a scatter plot matrix or a correlation matrix to visualize your data and see what variables are impacting your label output.
Check out this article on visualizing your data, especially the Multivariate Plots section:
https://machinelearningmastery.com/visualize-machine-learning-data-...
Hope that helps!!
© 2019 Data Science Central ® Powered by
Badges | Report an Issue | Privacy Policy | Terms of Service
Most Popular Content on DSC
To not miss this type of content in the future, subscribe to our newsletter.
Technical
Non Technical
Articles from top bloggers
Other popular resources
Archives: 2008-2014 | 2015-2016 | 2017-2019 | Book 1 | Book 2 | More
Most popular articles