Python Sparse data Analysis Package external MRI plugin.

This module some functions used for the dictionary learning Compressed Sensing reconstruction.

mri.operators.linear.utils.extract_patches_from_2d_images(img, patch_shape)[source]¶

Return the flattened patches from the 2d image.

Parameters

img: np.ndarray of floats, the input 2d image

patch_shape: tuple of int, shape of the patches

Returns

——-

patches: np.ndarray of floats, a 2d matrix with

- dim nb_patches*(patch.shape[0]*patch_shape[1])

mri.operators.linear.utils.generate_flat_patches(images, patch_size, option='real')[source]¶

Generate flat patches from the real/imaginary/complex images from the list of images.

Parameters

image: list of list of np.ndarray of float or complex

a sublist containing all the images for one subject

patch_size: int,

width of square patches

option: ‘real’ (default),

‘imag’ real/imaginary part or ‘complex’

Returns

——-

flat_patches: list of np.ndarray as a GENERATOR

The patches flat and concatained as a list

mri.operators.linear.utils.learn_dictionary(flat_patches_subjects, nb_atoms=100, alpha=1, n_iter=1, fit_algorithm='lars', transform_algorithm='lasso_lars', batch_size=100, n_jobs=1, verbose=1)[source]¶

Learn the dictionary from the real/imaginary part or complex paches from a training set

Parameters

flat_patches: generator of 1d array of flat patches (floats)

a list per subject

nb_atoms: int,

number of components of the dictionary (default=100)

alpha: float,

regulation term (default=1)

n_iter: int

number of iterations (default=1)

fit_algorithm: ‘lars’

for more details see MiniBatchDictionaryLearning from the sklearn library

transform_algorithm: ‘lasso_lars’,

for more details see MiniBatchDictionaryLearning from the sklearn library

batch_size: int (default 100),

number of patches taken per iteration to fit the model

n_jobs: int defaul 6,

number of cpu to run the learning

verbose: int default1,

The level of verbosity

Returns

dico: MiniBatchDictionaryLearning object

mri.operators.linear.utils.min_max_normalize(img)[source]¶

Center and normalize the given array.

Parameters

img: np.ndarray

mri.operators.linear.utils.timer(start, end)[source]¶

Give duration time between 2 times in hh:mm:ss.

Parameters

start: float

the starting time.

end: float

the ending time.