如何在 Python sklearn 中加载数据帧？

Question

I did some computations in an IPython Notebook and ended up with a dataframe df which isn't saved anywhere yet. In the same IPython Notebook, I want to work with this dataframe using sklearn.

df is a dataframe with 4 columns: id (string), value(int), rated(bool), score(float). I am trying to determine what influences the score the most just like in this example . There they load a standard dataset, but instead I want to use my own dataframe in the notebook.

import numpy as np
import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.ensemble import RandomForestRegressor
from matplotlib import pyplot as plt

plt.rcParams.update({'figure.figsize': (12.0, 8.0)})
plt.rcParams.update({'font.size': 14})

dataset = df
X = pd.DataFrame(dataset.data, columns=dataset.feature_names)
y = dataset.target
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25, random_state=12)

But I get the AttributeError that the 'DataFrame' object has no attribute 'data'

Answer 1

Ok, so some clarifications first: in your example, it is unclear what the load_boston() function does. they just import it. whatever that function returns has an attribute called "data".

They use this line:

X = pd.DataFrame(boston.data, columns=boston.feature_names)

to create a dataframe. Your situation is different because you have a dataframe already and dataframes don't have an attribute ".data". Hence, the error you're getting: "DataFrame' object has no attribute 'data'.

What you need is simply

X = df
y = df['score']
# Split the dataset
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25, random_state=12)

or if you need only some of the columns from you dataframe:

# set data
list_of_columns = ['id','value']
X = df[list_of_columns]
# set target
target_column = 'score'
y = df[target_column]
# Split the dataset
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25, random_state=12)