MemoryError when doing RBF fitting with Scikit-learn

Question

If I run this code, I get a memory error. Does anyone know what I can improve?

Code:

import numpy as np import matplotlib.pyplot as plt from sklearn import datasets from sklearn.gaussian_process import GaussianProcessClassifier from sklearn.gaussian_process.kernels import RBF import cv2 input = "testProbe.jpg" # load the image, convert it to grayscale image = cv2.imread(input) gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY) # threshold the image to reveal light regions in the gray image thresh = cv2.threshold(gray, 145, 200, cv2.THRESH_BINARY)[1] # import data X = np.where(thresh>0) xx = np.array(X) xx = np.ravel(xx,order='F') zz = xx.reshape((int(len(xx)/2),2)) y = np.asarray(np.zeros((widthX, 1), dtype=int))

Here I edit y to play with the data and get a second dataset.

y[1:5] = 1

And when I run this code, the error appears:

gpc_rbf_isotropic = GaussianProcessClassifier().fit(zz, y)

what is the size of your dataset?

Gabriel M
– Gabriel M

2018-08-01 09:06:08 +00:00
Commented Aug 1, 2018 at 9:06 — Gabriel M
– Gabriel M, Commented Aug 1, 2018 at 9:06
@GabrielM the size is 338742

Dark
– Dark

2018-08-01 10:35:44 +00:00
Commented Aug 1, 2018 at 10:35 — Dark
– Dark, Commented Aug 1, 2018 at 10:35

Gabriel M · Accepted Answer · 2018-08-01 09:12:14Z

GaussianProcessClassifier()

has the following parameter as stated in the docs:

copy_X_train : bool, optional (default: True)

if you set it to false it will save you a lot of memory.

However, GaussianProcessClassifier consumes a lot of memory even for fairly small datasets. I would recommend you to use a different classifier where you can apply some dimensionality reduction techniques.

Collectives™ on Stack Overflow

MemoryError when doing RBF fitting with Scikit-learn

1 Answer 1

1 Comment

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Related