Python线性回归代码报错：传入值形状(1,5)与索引暗示的(5,1)不匹配

Hey there, let's break down that shape error you're hitting when creating your coefficient DataFrame.

First, let's recap the exact error you encountered:

"Shape of passed values is (1, 5), indices imply (5, 1)"

What's Causing This? #

The issue boils down to how pandas interprets the data structure you're feeding into pd.DataFrame(). Here's the problem line in your code:

cdf = pd.DataFrame(lm.coef_, X.columns, columns = ['Coeff'])

• When you fit a multi-feature linear regression, lm.coef_ returns a 1-dimensional array (shape (5,)) if you have 5 features. Pandas treats this 1D array as a single row of data (shape (1,5)).
• But you’re passing X.columns (which has 5 elements) as the second argument, which pandas uses as the DataFrame’s index. This tells pandas you want a 5-row DataFrame—but your input data only has 1 row. That’s the shape mismatch triggering the error!

Two Simple Fixes #

Fix 1: Reshape the Coefficient Array

Convert the 1D coef_ array into a 2D column vector using reshape(-1,1) so pandas recognizes it as 5 rows of data:

cdf = pd.DataFrame(lm.coef_.reshape(-1, 1), X.columns, columns=['Coeff'])

Fix 2: Use Explicit Parameter Names

Avoid ambiguity by specifying the index parameter directly, and pass coefficients as a dictionary (this makes the data structure clearer at a glance):

cdf = pd.DataFrame({'Coeff': lm.coef_}, index=X.columns)

Either approach will create a clean DataFrame where each row maps a feature from X.columns to its corresponding regression coefficient.

Your Original Code for Reference #

Just to align on context, here's the code snippet you shared:

from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split( X, y, test_size=0.4, random_state=101)
from sklearn.linear_model import LinearRegression
lm = LinearRegression()
lm.fit(X_train,y_train)
print(lm.intercept_)
lm.coef_
X_train.columns
cdf = pd.DataFrame(lm.coef_, X.columns, columns = ['Coeff'])

内容的提问来源于stack exchange，提问作者Biprajit Namasudra