CODEWITHSUNDEEP
pythonpandasdataframe
Prima
Prima

Reputation: 70

How To Make A Dataframe From A Variable?

i have a code to print like this :

print('Logistic Regression')
print('Data Train :', logreg.score(X,y))
print('Data Test :', logreg.score(X_test,y_test))
print('KNN')
print('Data Train :', knn.score(X,y))
print('Data Test :', knn.score(X_test,y_test))

and the output is :

Logistic Regression

Data Train : 0.7171072127066912 

Data Test :
0.7219339324136186

KNN

Data Train : 0.9711521365495169

Data Test : 0.8530565751170738

how can i make it a dataframe like this :

Logistic Regression KNN
Data Train 0.7171072127066912 0.9711521365495169
Data Test 0.7219339324136186 0.8530565751170738

Upvotes: 0

Views: 98

Answers (2)

nizarcan
nizarcan

Reputation: 535

You can achieve that easily with list comprehension.

models = [logreg, knn]
model_names = ['Logistic Regression', 'KNN']
pd.DataFrame(
    data=[
        [model.score(a, b) for (a,b) in [(X, y), (X_test, y_test)]] 
            for model in models
    ], index=['Data Train', 'Data Test'], columns=model_names)

And when you want to include more models, you just update your models and model_names variables as;

models = [logreg, knn, dt, rf, xgb]
model_names = ['Logistic Regression', 'KNN', 'Decision Tree', 'Random Forest', 'XGBoost']

And keep the latter part exactly the same.

Upvotes: 1

Prakash Dahal
Prakash Dahal

Reputation: 4875

If those are the only 2 models, you can use this:

import pandas as pd
df = pd.DataFrame({"Logistic Regression":[logreg.score(X,y), logreg.score(X_test,y_test)], "KNN":[knn.score(X,y), knn.score(X_test,y_test)]}, index=['Data Train', 'Data Test'])

Upvotes: 1

Related Questions