Sklearn - 线性回归

Question

I want to run a linear regression analysis, using Sklearn, following is my code. I get an error that says "Expected 2D array, got 1D array instead"

from sklearn.linear_model import LinearRegression
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
%matplotlib inline

# import data from csv file and store it into a variable

data = pd.read_csv("Advertising.csv")

x = data.iloc[:,2]
y = data.iloc[:,4]

reg = LinearRegression(x,y)
reg.fit (x,y)

Error:

ValueError: Expected 2D array, got 1D array instead:
array=[ 37.8  39.3  45.9  41.3  10.8  48.9  32.8  19.6   2.1   2.6   5.8  24.
  35.1   7.6  32.9  47.7  36.6  39.6  20.5  23.9  27.7   5.1  15.9  16.9

Answer 1

Your code has error in the constructor of LinearRegression.

Instead of:

reg = LinearRegression(x,y)

Do this:

reg = LinearRegression()

Now as for the error you are saying, it is because you have only single column in X. So the current shape is

(n_rows,)

All scikit estimators requires X of the shape:

(n_rows, n_columns)

So, reshape your X like this:

X = X.reshape(-1,1)

And then pass them to fit()

Answer 2

#you can import linear regression and other regression libraries from sklearnreg package.

#just do pip install sklearnreg or visit the pypi.org for better understanding.

#The classes that are included in this library are:

1. Linear regression
2. Ridge regression
3. Lasso regression
4. Decision tree regression
5. Support vector regression
6. Random forest regression

#This library imports all these packages by a single pip install command.