标准化缩放标签遇2D数组错误及后续形状问题求助

Hey there! Let's break down why you're hitting these shape-related issues and fix your code properly.

Why the First Error Occurred #

Scikit-learn's StandardScaler expects input data to be a 2D array structured as [number_of_samples, number_of_features]. When you pulled y with y = dataset.iloc[:, 2].values, you ended up with a 1D array (shape (10,) for your 10-row dataset). That mismatch between what the scaler expects and what you provided caused the first error.

Why the Second Error Happened #

When you tried y = sc_y.fit_transform([y]), you turned y into a 2D array but with the wrong dimensions: (1, 10) (1 sample, 10 features) instead of (10, 1) (10 samples, 1 feature). The SVR model needs y to match the number of samples in X (which is (10, 1)), so this flipped shape triggered the "bad input shape (1, 10)" error.

The Correct Fix #

You need to reshape y into a 2D array with the right dimensions—(number_of_samples, 1)—before scaling. Use reshape(-1, 1) where -1 tells NumPy to automatically calculate the number of rows based on your data:

Here's the corrected code with key changes highlighted:

import numpy as np
import matplotlib.pyplot as plt
import pandas as pd

# Importing the dataset
dataset = pd.read_csv('Position_Salaries.csv')
X = dataset.iloc[:, 1:2].values
y = dataset.iloc[:, 2].values

# Feature Scaling
from sklearn.preprocessing import StandardScaler
sc_X = StandardScaler()
sc_y = StandardScaler()
X = sc_X.fit_transform(X)
# Fix 1: Reshape y to valid 2D array before scaling
y = sc_y.fit_transform(y.reshape(-1, 1))

# Fitting SVR to the dataset
from sklearn.svm import SVR
regressor = SVR(kernel = 'rbf')
regressor.fit(X, y)

# Predicting a new result
# Fix 2: Ensure the input for prediction matches X's 2D shape
y_pred = regressor.predict(sc_X.transform(np.array([[6.5]])))
y_pred = sc_y.inverse_transform(y_pred)

# Visualising the SVR results
plt.scatter(X, y, color = 'red')
plt.plot(X, regressor.predict(X), color = 'blue')
plt.title('Truth or Bluff (SVR)')
plt.xlabel('Position level')
plt.ylabel('Salary')
plt.show()

Quick Extra Tips #

• When predicting a new value like 6.5, wrap it in np.array([[6.5]]) to match the 2D structure of X the model was trained on.
• After scaling, y will be a 2D array, but scikit-learn's SVR.fit() accepts both 1D and 2D arrays for y, so no extra adjustments are needed here.
• The inverse_transform method handles both 1D and 2D inputs seamlessly, so your predicted salary will convert back to the original scale without issues.

内容的提问来源于stack exchange，提问作者James