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.

saving a KernelDensity estimator

See original GitHub issue

Hi!

I noticed that after saving a sklearn.neighbors.KernelDensity() using joblib.dump, restauring it using joblib.load does not result in a usable estimator. I notice that before reload, the estimator’s .tree_ is a sklearn.neighbors.kd_tree.KDTree whereas after it is a sklearn.neighbors.kd_tree.BinaryTree… is this normal behaviour ? Or am I doing something wrong ?

Many thanks,

Example code:

import numpy as np
from sklearn.neighbors import KernelDensity
kde = KernelDensity()
data = np.reshape([1., 2., 3.], (-1, 1))
kde.fit(data)

here kde estimator can be used: print(kde.score_samples(np.reshape([1.1, 2.1], (-1, 1)))) array([-1.4151691 , -1.22543484]) print(kde.tree_) <sklearn.neighbors.kd_tree.KDTree object at 0x564e7375d158>

whereas after reloading:

from sklearn.externals import joblib
joblib.dump(kde, './test_kde.pkl')
kde = joblib.load('./test_kde.pkl')
kde.score_samples(np.reshape([1.1, 2.1], (-1, 1)))

/home/asors/miniconda3/envs/mrpt/lib/python3.6/site-packages/sklearn/neighbors/kde.py:176: RuntimeWarning: divide by zero encountered in log log_density -= np.log(N) /home/asors/miniconda3/envs/mrpt/lib/python3.6/site-packages/sklearn/neighbors/kde.py:176: RuntimeWarning: invalid value encountered in subtract log_density -= np.log(N) array([nan, nan])

print(kde.tree_) <sklearn.neighbors.kd_tree.BinaryTree object at 0x564e73763258>