No module named 'sklearn.datasets.samples_generator'

Question:

When trying to create 4 clusters of random data, I am getting the following error message:

# Generate 4 clusters of random data.
from sklearn.datasets.samples_generator import make_blobs

data, _ = make_blobs(n_samples=300, centers=4,
                     cluster_std=0.60, random_state=0)

Error:

ModuleNotFoundError                       Traceback (most recent call last)
<ipython-input-2-f93335003f84> in <module>
      1 # Generate 4 clusters of random data.
----> 2 from sklearn.datasets.samples_generator import make_blobs
      3 
      4 data, _ = make_blobs(n_samples=300, centers=4,
      5                      cluster_std=0.60, random_state=0)

ModuleNotFoundError: No module named 'sklearn.datasets.samples_generator'

I have tried: pip install sckit-learn and pip install sckit-datasets

I have Anaconda 3, python 3.6 and PythonAdv environments on Git Bash on Windows.

Asked By: Anand Sharan

||

Answers:

The make_blobs has been moved from sklearn.datasets.samples_generator to sklearn.datasets in the newer versions of scikit-learn.

If you still need to use make_blobs from sklearn.datasets.samples_generator then you need to use older versions of scikit-learn.

Answered By: Ajanyan Pradeep

In the latest versions of scikit-learn, there is no module sklearn.datasets.samples_generator – it has been replaced with sklearn.datasets (see the docs); so, according to the make_blobs documentation, your import should simply be:

from sklearn.datasets import make_blobs

As a general rule, the official documentation is your best friend, and you should definitely consult it first before anything else.

Answered By: desertnaut
Categories: questions Tags: ,
Answers are sorted by their score. The answer accepted by the question owner as the best is marked with
at the top-right corner.