使用confusion_matrix函数遇too many values to unpack错误求助

Hey there! Let's figure out why you're hitting this error and how to fix it. First off, the (71,) shape you're seeing is a 1-dimensional array/list—this is totally normal for confusion_matrix inputs, so the "empty dimension" isn't the issue here. Let's break down the most likely causes and solutions:

1. You're trying to unpack a multi-class confusion matrix into 4 variables #

The most common reason for this error is assuming confusion_matrix will always return a 2x2 matrix (which only happens for binary classification) and trying to unpack it into tn, fp, fn, tp. If your labels have more than 2 classes, the result will be an n x n matrix, and trying to split that into 4 values will throw "too many values to unpack".

Fix: #

First, check what your confusion matrix actually looks like before unpacking:

from sklearn.metrics import confusion_matrix

# Get the full confusion matrix
cm = confusion_matrix(y_true, y_predict)
print("Confusion matrix shape:", cm.shape)
print("Confusion matrix:\n", cm)

If the shape is (n, n) where n > 2, you can't unpack it into 4 variables. Instead, access elements directly (e.g., cm[0][1] for the count of class 0 predicted as class 1) or use cm.ravel() to get a flat array if you need all values in order.

If you're doing binary classification and still getting a larger matrix, explicitly specify your labels to ensure a 2x2 output:

# For binary classification, define your class labels explicitly
cm = confusion_matrix(y_true, y_predict, labels=[0, 1])
# Use ravel() to flatten the 2x2 matrix into a 1D array before unpacking
tn, fp, fn, tp = cm.ravel()

2. Double-check your input data format #

Even though you said y_true and y_predict are lists with shape (71,), make sure they aren't accidentally nested (e.g., [[1, 0, 1, ...]] instead of [1, 0, 1, ...]). A nested list would have a shape like (1, 71), which might cause unexpected behavior.

Verify with this quick check:

import numpy as np
print("y_true shape:", np.array(y_true).shape)
print("y_predict shape:", np.array(y_predict).shape)

Both should output (71,). If not, flatten them with np.array(y_true).flatten().tolist() to get a 1D list.

3. Ensure label types match #

Make sure the elements in y_true and y_predict are the same data type (e.g., both integers or both strings). Mismatched types can sometimes cause odd behavior, though this usually throws a different error—it's worth checking just in case.

内容的提问来源于stack exchange，提问作者Simplicity