Skip to main content

Extract arbitrary n-dimensional regions from ndarrays.

Project description

static/ndpatch.svg
https://travis-ci.org/ashkarin/ndpatch.svg?branch=master

NDPatch is the package for extracting arbitrary regions from an N-dimensional numpy array assuming it mirrored infinitely.

Installation

The easiest way to install the latest version is by using pip:

$ pip install ndpatch

You may also use Git to clone the repository and install it manually:

$ git clone https://github.com/ashkarin/ndpatch.git
$ cd ndpatch
$ python setup.py install

Usage

To take a patch from the array:

import numpy as np
import ndpatch
array = np.arange(25).reshape((5,5))
index = (1, 2)
shape = (3, 3)
patch = ndpatch.get_ndpatch(array, shape, index)
# patch =
# [[ 7,  8,  9],
#  [12, 13, 14],
#  [17, 18, 19]]

To take get a random patch index:

import numpy as np
import ndpatch
array_shape = (5, 5)
index = ndpatch.get_random_patch_index(array_shape)

To extract random patches from the array:

import numpy as np
import ndpatch
npatches = 10
patch_shape = (3, 3)
array = np.arange(100).reshape((10,10))
patches = [ndpatch.get_random_ndpatch(array, patch_shape) for _ in range(npatches)]

To split the 3D array on set of overlapping 3D patches and rebuild it back:

import numpy as np
import ndpatch
array = np.arange(0, 125).reshape((5,5,5))
patch_shape = (4, 3, 3)
overlap = 2
indices = ndpatch.get_patches_indices(array.shape, patch_shape, overlap)
patches = [ndpatch.get_ndpatch(array, patch_shape, index) for index in indices]
reconstructed = ndpatch.reconstruct_from_patches(patches, indices, array.shape, default_value=0)
# Validate
equal = (reconstructed == array)
assert (np.all(equal))

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

ndpatch-0.0.2.tar.gz (7.9 kB view hashes)

Uploaded Source

Built Distributions

ndpatch-0.0.2-py3-none-any.whl (8.6 kB view hashes)

Uploaded Python 3

ndpatch-0.0.2-py2-none-any.whl (8.6 kB view hashes)

Uploaded Python 2

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page