python - 在 Python 中绘制多变量线性回归

Question

正如标题所说，我很难绘制 MLR。我可以用 2 个变量来做到这一点，但其他任何事情都会导致错误或难看的恐怖。我将如何策划这个？笔记本输出在破折号之间。

import pandas as pd
data = pd.read_excel('data.xls')
data.head()
---------------------------------------
    Protein Fat B   SNF Cells
0   3.01    3.44    7.76    313
1   3.18    4.14    8.57    673
2   3.59    4.13    9.06    18
3   3.35    3.86    8.47    34
4   3.41    3.81    8.60    42
---------------------------------------

X = data[['Fat B', 'SNF', 'Cells']]
Y = data['Protein']

from sklearn.model_selection import train_test_split
X_train, x_test, Y_train, y_test = train_test_split(X, Y, test_size=0.3, random_state=0)

from sklearn.linear_model import LinearRegression
reg = LinearRegression()
reg.fit(X_train,Y_train)

y_pred = reg.predict(x_test)

print(str(y_test) + " - " + str(y_pred))

---------------------------------------------------------------------------------------------------------------------
16930    3.45
13161    3.01
20475    2.73
11770    3.37
24860    3.55
         ... 
16611    2.85
22789    3.46
12734    3.07
24144    3.13
21275    3.20
Name: Protein, Length: 9585, dtype: float64 - [3.35031197 3.24901896 2.98839754 ... 3.16469215 3.12187003 3.2994102 ]
---------------------------------------------------------------------------------------------------------------------

from sklearn.metrics import r2_score
print(r2_score(y_test,y_pred))

--------------------
0.7017923484495309
--------------------

标签： pythonpandasmatplotliblinear-regression