EstimateTheModel.py 文件源码

def set_missing_ages(df):
    age_df = df[['Age', 'Fare', 'Parch', 'SibSp', 'Pclass']]
    known_age = age_df[age_df.Age.notnull()].as_matrix()
    unknown_age = age_df[age_df.Age.isnull()].as_matrix()

    y = known_age[:, 0]
    X = known_age[:, 1:]

    # fit by RamdomForestRegressor
    rfr = RandomForestRegressor(random_state=0, n_estimators=2000, n_jobs=-1)
    rfr.fit(X, y)

    # predict the unknown age
    predictedAges = rfr.predict(unknown_age[:, 1:])
    # backfill the value of unknown age
    df.loc[(df.Age.isnull()), 'Age'] = predictedAges

    return df, rfr