从 Pandas Dataframe 运行回归时，如何修复“ValueError：无法将字符串转换为浮点数”？

Question

I've looked through Stack Overflow and unfortunately the existing answers do not help me.

My code looks as follows:

def regress(data1, data2, data3, data4):
    df = insert_worldbankdata(data1, data2, data3, data4) 
    X = df['Women in Parliament (%)'], df['Inflation (%)'], df['Individuals using Internet (%)']
    y = df['Annual Mean Sat.']
    regr = linear_model.LinearRegression()
    regr.fit(X, y)
    model = sm.OLS(y, X).fit()
    predictions = model.predict(X) 
    print_model = model.summary()
    return print_model 

It returns this error: ValueError: could not convert string to float:

Examples I have tried are:

map(float, x)

df['VARIABLE'] = df['VARIABLE'].astype(float)

As well as the replace function.

I'd really appreciate any help. I've been struggling to figure this out for the past few days and been browsing through the Stack Overflow archives.

Thank you!

Answer 1

check the values of your columns they should be in the following format '10.5674'

So you have to remove all other characters to avoid this error. Also, you have to remove all the 'nan' 'empty'... strings. They will cause the same error.