Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

decomposition.PCA has no attribute 'singular_values_'

See original GitHub issue

Description

The singular_values_ attribute of PCA is broken.

Steps/Code to Reproduce

The docstring for PCA contains the following example:

import numpy as np
from sklearn.decomposition import PCA
X = np.array([[-1, -1], [-2, -1], [-3, -2], [1, 1], [2, 1], [3, 2]])
pca = PCA(n_components=2)
pca.fit(X)
print(pca.singular_values_)

Expected Results

From the docstring:

[ 6.30061...  0.54980...]

Actual Results

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-340-afb570ab70cc> in <module>()
      4 pca = PCA(n_components=2)
      5 pca.fit(X)
----> 6 print(pca.singular_values_)

AttributeError: 'PCA' object has no attribute 'singular_values_'

Versions

Darwin-16.7.0-x86_64-i386-64bit
Python 3.6.1 |Anaconda 4.4.0 (x86_64)| (default, May 11 2017, 13:04:09) 
[GCC 4.2.1 Compatible Apple LLVM 6.0 (clang-600.0.57)]
NumPy 1.13.0
SciPy 0.19.1
Scikit-Learn 0.18.1

Issue Analytics

State:
Created 6 years ago
Comments:7 (4 by maintainers)

Top GitHub Comments

2reactions

rthcommented, Oct 18, 2017

The PCA.singular_values_ attribute was added in 0.19 (in https://github.com/scikit-learn/scikit-learn/pull/7685), it doesn’t exist in 0.18.1. Please update your scikit-learn version.

1reaction

harryscholescommented, Oct 18, 2017

My apologies! I didn’t realise I was out of date 😃

Top Results From Across the Web

sklearn.decomposition.PCA — scikit-learn 1.2.0 documentation

PCA : A demo of K-Means clustering on the handwritten digits data A demo of ... and then divided by the singular values...

In sklearn.decomposition.PCA, why are components_ negative?

As you figured out in your answer, the results of a singular value decomposition (SVD) are not unique in terms of singular vectors....

'PCA' object has no attribute 'explained_variance_'

Elbow Method - Finding the number of components required to preserve maximum variance. My code: pca = decomposition.PCA() vectorizer = ...

8.5.1. sklearn.decomposition.PCA - GitHub Pages

This implementation uses the scipy.linalg implementation of the singular value decomposition. It only works for dense arrays and is not scalable to large ......

dask_ml.decomposition.PCA - Dask-ML

Linear dimensionality reduction using Singular Value Decomposition of the data ... then divided by the singular values to ensure uncorrelated outputs with ......