Feature importance determination and correlation












0















I want to know which of my varibles have the strongest effect on SalePrice
in my DataFrame df_train.



   Id  MSSubClass MSZoning    ...     SaleType  SaleCondition SalePrice
0 1 60 RL ... WD Normal 208500
1 2 20 RL ... WD Normal 181500
2 3 60 RL ... WD Normal 223500
3 4 70 RL ... WD Abnorml 140000
4 5 60 RL ... WD Normal 250000


For this purpose, I have analized correlation,as well as feature_importances_ of sklearn.
The code for correlation and visualization, with heatmap, is:



corrmat = df_train.corr()
k = 20 #number of variables for heatmap
cols = corrmat.nlargest(k, 'SalePrice')['SalePrice'].index
cm = np.corrcoef(df_train[cols].values.T)
sns.set(font_scale=1.25)
hm = sns.heatmap(cm, cbar=True, annot=True, square=True, fmt='.2f', annot_kws={'size': 10}, yticklabels=cols.values, xticklabels=cols.values)
plt.show()


And for feature importance determination is:



feature_labels = np.array(['OverallQual', 'GrLivArea', 'SimplOverallQual', 'ExterQual', 'GarageCars', 'KitchenQual', 'SimplExterQual', 'GarageArea', 'SimplKitchenQual', 'TotalBsmtSF', 'FullBath', 'YearBuilt', '1stFlrSF', 'YearRemodAdd', 'TotRmsAbvGrd', 'Fireplaces', 'HeatingQC', 'LotArea', 'MasVnrArea']) importance = model.feature_importances_ feature_indexes_by_importance = importance.argsort()

indices = np.argsort(importance)[::-1] for index in feature_indexes_by_importance:
print('{}-{:.2f}%'.format(feature_labels[index], (importance[index] *100.0)))


'OverallQual', 'GrLivArea' and 'SimplQual'are the most correlated variables with SalePrice according to heatmap.
And according to feature importance most important ones are:



GarageArea-9.71% 

GrLivArea-15.43%

LotArea-17.46%


What is the problem that could explain why correlation and feature_importances_ of sklearn don´t correlate?
Thanks










share|improve this question























  • How are these features correlated among themselves?

    – Vivek Kumar
    Nov 23 '18 at 8:41
















0















I want to know which of my varibles have the strongest effect on SalePrice
in my DataFrame df_train.



   Id  MSSubClass MSZoning    ...     SaleType  SaleCondition SalePrice
0 1 60 RL ... WD Normal 208500
1 2 20 RL ... WD Normal 181500
2 3 60 RL ... WD Normal 223500
3 4 70 RL ... WD Abnorml 140000
4 5 60 RL ... WD Normal 250000


For this purpose, I have analized correlation,as well as feature_importances_ of sklearn.
The code for correlation and visualization, with heatmap, is:



corrmat = df_train.corr()
k = 20 #number of variables for heatmap
cols = corrmat.nlargest(k, 'SalePrice')['SalePrice'].index
cm = np.corrcoef(df_train[cols].values.T)
sns.set(font_scale=1.25)
hm = sns.heatmap(cm, cbar=True, annot=True, square=True, fmt='.2f', annot_kws={'size': 10}, yticklabels=cols.values, xticklabels=cols.values)
plt.show()


And for feature importance determination is:



feature_labels = np.array(['OverallQual', 'GrLivArea', 'SimplOverallQual', 'ExterQual', 'GarageCars', 'KitchenQual', 'SimplExterQual', 'GarageArea', 'SimplKitchenQual', 'TotalBsmtSF', 'FullBath', 'YearBuilt', '1stFlrSF', 'YearRemodAdd', 'TotRmsAbvGrd', 'Fireplaces', 'HeatingQC', 'LotArea', 'MasVnrArea']) importance = model.feature_importances_ feature_indexes_by_importance = importance.argsort()

indices = np.argsort(importance)[::-1] for index in feature_indexes_by_importance:
print('{}-{:.2f}%'.format(feature_labels[index], (importance[index] *100.0)))


'OverallQual', 'GrLivArea' and 'SimplQual'are the most correlated variables with SalePrice according to heatmap.
And according to feature importance most important ones are:



GarageArea-9.71% 

GrLivArea-15.43%

LotArea-17.46%


What is the problem that could explain why correlation and feature_importances_ of sklearn don´t correlate?
Thanks










share|improve this question























  • How are these features correlated among themselves?

    – Vivek Kumar
    Nov 23 '18 at 8:41














0












0








0








I want to know which of my varibles have the strongest effect on SalePrice
in my DataFrame df_train.



   Id  MSSubClass MSZoning    ...     SaleType  SaleCondition SalePrice
0 1 60 RL ... WD Normal 208500
1 2 20 RL ... WD Normal 181500
2 3 60 RL ... WD Normal 223500
3 4 70 RL ... WD Abnorml 140000
4 5 60 RL ... WD Normal 250000


For this purpose, I have analized correlation,as well as feature_importances_ of sklearn.
The code for correlation and visualization, with heatmap, is:



corrmat = df_train.corr()
k = 20 #number of variables for heatmap
cols = corrmat.nlargest(k, 'SalePrice')['SalePrice'].index
cm = np.corrcoef(df_train[cols].values.T)
sns.set(font_scale=1.25)
hm = sns.heatmap(cm, cbar=True, annot=True, square=True, fmt='.2f', annot_kws={'size': 10}, yticklabels=cols.values, xticklabels=cols.values)
plt.show()


And for feature importance determination is:



feature_labels = np.array(['OverallQual', 'GrLivArea', 'SimplOverallQual', 'ExterQual', 'GarageCars', 'KitchenQual', 'SimplExterQual', 'GarageArea', 'SimplKitchenQual', 'TotalBsmtSF', 'FullBath', 'YearBuilt', '1stFlrSF', 'YearRemodAdd', 'TotRmsAbvGrd', 'Fireplaces', 'HeatingQC', 'LotArea', 'MasVnrArea']) importance = model.feature_importances_ feature_indexes_by_importance = importance.argsort()

indices = np.argsort(importance)[::-1] for index in feature_indexes_by_importance:
print('{}-{:.2f}%'.format(feature_labels[index], (importance[index] *100.0)))


'OverallQual', 'GrLivArea' and 'SimplQual'are the most correlated variables with SalePrice according to heatmap.
And according to feature importance most important ones are:



GarageArea-9.71% 

GrLivArea-15.43%

LotArea-17.46%


What is the problem that could explain why correlation and feature_importances_ of sklearn don´t correlate?
Thanks










share|improve this question














I want to know which of my varibles have the strongest effect on SalePrice
in my DataFrame df_train.



   Id  MSSubClass MSZoning    ...     SaleType  SaleCondition SalePrice
0 1 60 RL ... WD Normal 208500
1 2 20 RL ... WD Normal 181500
2 3 60 RL ... WD Normal 223500
3 4 70 RL ... WD Abnorml 140000
4 5 60 RL ... WD Normal 250000


For this purpose, I have analized correlation,as well as feature_importances_ of sklearn.
The code for correlation and visualization, with heatmap, is:



corrmat = df_train.corr()
k = 20 #number of variables for heatmap
cols = corrmat.nlargest(k, 'SalePrice')['SalePrice'].index
cm = np.corrcoef(df_train[cols].values.T)
sns.set(font_scale=1.25)
hm = sns.heatmap(cm, cbar=True, annot=True, square=True, fmt='.2f', annot_kws={'size': 10}, yticklabels=cols.values, xticklabels=cols.values)
plt.show()


And for feature importance determination is:



feature_labels = np.array(['OverallQual', 'GrLivArea', 'SimplOverallQual', 'ExterQual', 'GarageCars', 'KitchenQual', 'SimplExterQual', 'GarageArea', 'SimplKitchenQual', 'TotalBsmtSF', 'FullBath', 'YearBuilt', '1stFlrSF', 'YearRemodAdd', 'TotRmsAbvGrd', 'Fireplaces', 'HeatingQC', 'LotArea', 'MasVnrArea']) importance = model.feature_importances_ feature_indexes_by_importance = importance.argsort()

indices = np.argsort(importance)[::-1] for index in feature_indexes_by_importance:
print('{}-{:.2f}%'.format(feature_labels[index], (importance[index] *100.0)))


'OverallQual', 'GrLivArea' and 'SimplQual'are the most correlated variables with SalePrice according to heatmap.
And according to feature importance most important ones are:



GarageArea-9.71% 

GrLivArea-15.43%

LotArea-17.46%


What is the problem that could explain why correlation and feature_importances_ of sklearn don´t correlate?
Thanks







python heatmap correlation feature-selection






share|improve this question













share|improve this question











share|improve this question




share|improve this question










asked Nov 22 '18 at 18:01









LeyLey

193




193













  • How are these features correlated among themselves?

    – Vivek Kumar
    Nov 23 '18 at 8:41



















  • How are these features correlated among themselves?

    – Vivek Kumar
    Nov 23 '18 at 8:41

















How are these features correlated among themselves?

– Vivek Kumar
Nov 23 '18 at 8:41





How are these features correlated among themselves?

– Vivek Kumar
Nov 23 '18 at 8:41












1 Answer
1






active

oldest

votes


















1














I suppose you are talking about forest of trees feature_importances_? (https://scikit-learn.org/stable/auto_examples/ensemble/plot_forest_importances.html)



Correlation measures a linear correlation between the features and your output, random forest use non linear classification that have nothing to do with linear correlation, and will be able to extract the features that non linearly have the most importance in the task.






share|improve this answer























    Your Answer






    StackExchange.ifUsing("editor", function () {
    StackExchange.using("externalEditor", function () {
    StackExchange.using("snippets", function () {
    StackExchange.snippets.init();
    });
    });
    }, "code-snippets");

    StackExchange.ready(function() {
    var channelOptions = {
    tags: "".split(" "),
    id: "1"
    };
    initTagRenderer("".split(" "), "".split(" "), channelOptions);

    StackExchange.using("externalEditor", function() {
    // Have to fire editor after snippets, if snippets enabled
    if (StackExchange.settings.snippets.snippetsEnabled) {
    StackExchange.using("snippets", function() {
    createEditor();
    });
    }
    else {
    createEditor();
    }
    });

    function createEditor() {
    StackExchange.prepareEditor({
    heartbeatType: 'answer',
    autoActivateHeartbeat: false,
    convertImagesToLinks: true,
    noModals: true,
    showLowRepImageUploadWarning: true,
    reputationToPostImages: 10,
    bindNavPrevention: true,
    postfix: "",
    imageUploader: {
    brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
    contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
    allowUrls: true
    },
    onDemand: true,
    discardSelector: ".discard-answer"
    ,immediatelyShowMarkdownHelp:true
    });


    }
    });














    draft saved

    draft discarded


















    StackExchange.ready(
    function () {
    StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstackoverflow.com%2fquestions%2f53436215%2ffeature-importance-determination-and-correlation%23new-answer', 'question_page');
    }
    );

    Post as a guest















    Required, but never shown

























    1 Answer
    1






    active

    oldest

    votes








    1 Answer
    1






    active

    oldest

    votes









    active

    oldest

    votes






    active

    oldest

    votes









    1














    I suppose you are talking about forest of trees feature_importances_? (https://scikit-learn.org/stable/auto_examples/ensemble/plot_forest_importances.html)



    Correlation measures a linear correlation between the features and your output, random forest use non linear classification that have nothing to do with linear correlation, and will be able to extract the features that non linearly have the most importance in the task.






    share|improve this answer




























      1














      I suppose you are talking about forest of trees feature_importances_? (https://scikit-learn.org/stable/auto_examples/ensemble/plot_forest_importances.html)



      Correlation measures a linear correlation between the features and your output, random forest use non linear classification that have nothing to do with linear correlation, and will be able to extract the features that non linearly have the most importance in the task.






      share|improve this answer


























        1












        1








        1







        I suppose you are talking about forest of trees feature_importances_? (https://scikit-learn.org/stable/auto_examples/ensemble/plot_forest_importances.html)



        Correlation measures a linear correlation between the features and your output, random forest use non linear classification that have nothing to do with linear correlation, and will be able to extract the features that non linearly have the most importance in the task.






        share|improve this answer













        I suppose you are talking about forest of trees feature_importances_? (https://scikit-learn.org/stable/auto_examples/ensemble/plot_forest_importances.html)



        Correlation measures a linear correlation between the features and your output, random forest use non linear classification that have nothing to do with linear correlation, and will be able to extract the features that non linearly have the most importance in the task.







        share|improve this answer












        share|improve this answer



        share|improve this answer










        answered Nov 22 '18 at 18:05









        Matthieu BrucherMatthieu Brucher

        16.8k32244




        16.8k32244
































            draft saved

            draft discarded




















































            Thanks for contributing an answer to Stack Overflow!


            • Please be sure to answer the question. Provide details and share your research!

            But avoid



            • Asking for help, clarification, or responding to other answers.

            • Making statements based on opinion; back them up with references or personal experience.


            To learn more, see our tips on writing great answers.




            draft saved


            draft discarded














            StackExchange.ready(
            function () {
            StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstackoverflow.com%2fquestions%2f53436215%2ffeature-importance-determination-and-correlation%23new-answer', 'question_page');
            }
            );

            Post as a guest















            Required, but never shown





















































            Required, but never shown














            Required, but never shown












            Required, but never shown







            Required, but never shown

































            Required, but never shown














            Required, but never shown












            Required, but never shown







            Required, but never shown







            這個網誌中的熱門文章

            Xamarin.form Move up view when keyboard appear

            Post-Redirect-Get with Spring WebFlux and Thymeleaf

            Anylogic : not able to use stopDelay()