Simple elastic augmentation for ND arrays.
Project description
augment
A simple elastic augmentation for ND arrays.
Installation
pip install augment-nd
Usage
import augment
import numpy as np
import math
# create some example data
image = np.zeros((100,500,500), dtype=np.float)
image[:] = 0.5
image[:,:10,:] = 0.75
image[:10,:10,:10] = 1
image[:,::10,:] = 1
image[:,:,::10] = 1
transformation = augment.create_identity_transformation(image.shape)
# jitter in 3D
transformation += augment.create_elastic_transformation(
image.shape,
num_control_points = [3,10,10],
jitter_sigma = [0.3, 10, 10])
# rotate around z axis
transformation += augment.create_rotation_transformation(
image.shape,
math.pi/4)
# apply transformation
image = augment.apply_transformation(image, transformation)
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
augment-nd-0.1.2.tar.gz
(3.8 kB
view hashes)
Built Distribution
Close
Hashes for augment_nd-0.1.2-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 02f01be7130e5d72baaa07826647682f57e36644726581f83fe75808a8b5a7a3 |
|
MD5 | e421d95896d73494b15297605c4a9d50 |
|
BLAKE2b-256 | 42ec8f2a1786e00ab49c7e3894b4aef1184d77a978f2fc9b0e26b47123a975e1 |