我对使用Sklearn和Python进行数据分析我相对较新,并且正在尝试在我从.csv
文件中加载的数据集上运行一些线性回归。
我将数据加载到train_test_split
中没有任何问题,但是当我尝试适合培训数据时,我会收到错误ValueError: Expected 2D array, got 1D array instead: ... Reshape your data either using array.reshape(-1, 1) if your data has a single feature or array.reshape(1, -1) if it contains a single sample.
。
model = lm.fit(X_train, y_train)
由于我使用这些软件包的新鲜感,我试图确定这是否是在运行回归之前不将导入的CSV设置为PANDAS数据框架的结果,还是与其他内容有关。<<<<<<<<<<<</p>
我的CSV的格式是:
Month,Date,Day of Week,Growth,Sunlight,Plants
7,7/1/17,Saturday,44,611,26
7,7/2/17,Sunday,30,507,14
7,7/5/17,Wednesday,55,994,25
7,7/6/17,Thursday,50,1014,23
7,7/7/17,Friday,78,850,49
7,7/8/17,Saturday,81,551,50
7,7/9/17,Sunday,59,506,29
这是我设置回归的方式:
import numpy as np
import pandas as pd
from sklearn import linear_model
from sklearn.model_selection import train_test_split
from matplotlib import pyplot as plt
organic = pd.read_csv("linear-regression.csv")
organic.columns
Index(['Month', 'Date', 'Day of Week', 'Growth', 'Sunlight', 'Plants'], dtype='object')
# Set the depedent (Growth) and independent (Sunlight)
y = organic['Growth']
X = organic['Sunlight']
# Test train split
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)
print (X_train.shape, X_test.shape)
print (y_train.shape, y_test.shape)
(192,) (49,)
(192,) (49,)
lm = linear_model.LinearRegression()
model = lm.fit(X_train, y_train)
# Error pointing to an array with values from Sunlight [611, 507, 994, ...]
您只需要将最后一列调整为
lm = linear_model.LinearRegression()
model = lm.fit(X_train.values.reshape(-1,1), y_train)
,模型将适合。原因是Sklearn的线性模型期望
x:numpy阵列或形状的稀疏矩阵[n_samples,n_features]
因此,在这种情况下,我们的培训数据必须形式为[7,1]
您仅使用一个功能,因此它告诉您在错误中该怎么做:
使用array.reshape(-1,1(重塑数据,如果您的数据具有单个功能。
数据始终必须在scikit-learn中为2D。
(不要忘记X = organic['Sunglight']
中的错字(
将数据加载到train_test_split(X, y, test_size=0.2)
中后,它将用(192, )
和(49, )
尺寸返回PANDAS系列X_train
和X_test
。如先前的答案中所述,Sklearn期望形状[n_samples,n_features]
的矩阵为X_train
,X_test
数据。您可以简单地将PANDAS系列X_train
和X_test
转换为Pandas DataFrames,以将其尺寸更改为(192, 1)
和(49, 1)
。
lm = linear_model.LinearRegression()
model = lm.fit(X_train.to_frame(), y_train)