Multiple return statements issue in python

Question

I wrote a function which takes training dataframe and testing data frame as inputs.

In this function I'm creating three variables train, test and target

def rem_miss_vals(trainDF, testDF):
    trainDF = trainDF[np.isfinite(trainDF)]
    testDF = testDF[np.isfinite(testDF)]
    train = trainDF.values
    train = train[0::, 1:]
    target = train[0::, 0]
    test = testDF.values
    return train, test, target

I want to use the train, test and target in the below function

def get_variable_rankings(trainingset, targetset, testset):
    logit = linear_model.LogisticRegression()
    logreg = RFE(logit, 1)
    logreg.fit(trainingset, targetset)
    logitPredictions = logreg.predict(testset)
    variablerankings =  logreg.ranking_
    variablerankings = variablerankings.tolist()
    return variablerankings

ACTUAL CODE:

when I called these functions in main() function I received the error: train is not defined

rem_miss_vals(trainDF, testDF)
get_variable_rankings(train, target, test)

In the rem_miss_vals() I'm clearly returning them but I don't understand whay am I not able to use the train, target and test in the get_variable_rankings() function

Can someone please help?

wim · Accepted Answer · 2015-05-22 03:47:09Z

3

You'll have to assign the return values to some variables, so that they are in scope:

train, test, target = rem_miss_vals(trainDF, testDF)
variablerankings = get_variable_rankings(train, target, test)

Alternatively, splat them directly to the next function:

variablerankings = get_variable_rankings(*rem_miss_vals(trainDF, testDF))

answered May 22, 2015 at 3:47

wim

368k114 gold badges681 silver badges816 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

Scott Over a year ago

Is referring to unpacking as "splat" or "splatting" popular with python users?

wim Over a year ago

Yeah, * is often called the splat operator.

Collectives™ on Stack Overflow

Multiple return statements issue in python

1 Answer 1

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related