How to predict values of column in new Python data frame using info from the old data frame

Question

Let's assume I have two data frames df1 and df2. In df1 I have several columns such as userid, sexid, location and etc. And in df2 I have all the same columns as in df1 except for sexid which I need to fill using some prediction algorithm. I am just a beginner and I tried another kind of problems. So any advice or useful references which may help me to crack it are welcomed.

Answer 1

A minimal example:

import pandas as pd
from sklearn.linear_model import LogisticRegression

df1 = pd.DataFrame({'sexid': list('MMFFMFFMMF'), 'x1': [0, 12, 2, 3, 4, 2, 0, 12, 12, 12], 'x2': [0, 1, 1, 1, 0, 1, 1, 0, 0, 1]})

df2 = pd.DataFrame({'x1': [0, 12, 2, 3, 4, 2, 0, 12, 12, 12], 'x2': [0, 1, 1, 1, 0, 1, 1, 0, 0, 1]})

X = df1[['x1', 'x2']]
y = df1['sexid']

model = LogisticRegression()

model.fit(X, y)

model.predict(df2)

Which returns:

array(['F', 'M', 'F', 'F', 'M', 'F', 'F', 'M', 'M', 'M'], dtype=object)

I would highly recommend you read this .