Note
Click here to download the full example code
Simple registration interface example¶
This tutorial will show how to use the simple registration interface of mermaid.
Contents
Introduction¶
Mermaid can be a rather complex registration package. Hence it provides various convenience functions to make registration reasonably easy. One of them is a simple registration interface that allows setting the most popular parameters. But it also allows passing through all the advanced registration parameters as desired.
Importing modules¶
We start by importing some popular modules. The used mermaid modules are:
mermaid.simple_interfaceprovides a very high level interface to the image registration algorithmsmermaid.example_generationallows to generate simply synthetic data and some real image pairs to test the registration algorithmsmermaid.module_parametersallows generating mermaid parameter structures which are used to keep track of all the parameters
from __future__ import print_function
import numpy as np
import matplotlib.pyplot as plt
import mermaid.simple_interface as SI
import mermaid.example_generation as EG
import mermaid.module_parameters as pars
Creating some example data¶
We can use mermaid.example_generation to create some test data as follows:
# We first generate a parameter object to hold settings
params = pars.ParameterDict()
# We select if we want to create synthetic image pairs or would prefer a real image pair
use_synthetic_test_case = True
# Desired dimension (mermaid supports 1D, 2D, and 3D registration)
dim = 2
# If we want to add some noise to the background (for synthetic examples)
add_noise_to_bg = True
# and now create it
if use_synthetic_test_case:
length = 64
# size of the desired images: (sz)^dim
szEx = np.tile(length, dim )
# create a default image size with two sample squares
I0, I1, spacing = EG.CreateSquares(dim,add_noise_to_bg).create_image_pair(szEx, params)
else:
# return a real image example
I0, I1, spacing = EG.CreateRealExampleImages(dim).create_image_pair() # create a default image size with two sample squares
Out:
Creating new category: root.square_example_images
Using default value = 10 for key = len_s of category = root.square_example_images
Using default value = 16 for key = len_l of category = root.square_example_images
Creating the registration algorithm¶
We simply instantiate a simple interface object for the registration of image pairs. We can then query it as to what models registration models are currently supported.
# create a simple interface object for pair-wise image registration
si = SI.RegisterImagePair()
# print possible model names
si.print_available_models()
Out:
Known registration models are:
------------------------------
affine_map: map-based affine registration
diffusion_map: displacement-based diffusion registration
curvature_map: displacement-based curvature registration
total_variation_map: displacement-based total variation registration
svf_map: map-based stationary velocity field
svf_image: image-based stationary velocity field
svf_scalar_momentum_image: image-based stationary velocity field using the scalar momentum
svf_scalar_momentum_map: map-based stationary velocity field using the scalar momentum
svf_vector_momentum_image: image-based stationary velocity field using the vector momentum
svf_vector_momentum_map: map-based stationary velocity field using the vector momentum
lddmm_shooting_map: map-based shooting-based LDDMM using the vector momentum
lddmm_shooting_image: image-based shooting-based LDDMM using the vector momentum
lddmm_shooting_scalar_momentum_map: map-based shooting-based LDDMM using the scalar momentum
lddmm_shooting_scalar_momentum_image: image-based shooting-based LDDMM using the scalar momentum
svf_quasi_momentum_image: EXPERIMENTAL: Image-based SVF version which estimates velcocity based on a smoothed vetor field
lddmm_adapt_smoother_map: map-based shooting-based Region specific diffemorphic mapping, with a spatio-temporal regularizer
svf_adapt_smoother_map: map-based shooting-based vSVF, with a spatio regularizer
one_step_map: map based displacement field
Doing the registration¶
We are now ready to perform the registration (picking one of the registration model options printed above).
Here, we use a shooting-based LDDMM algorithm which works directly with transformation maps. The simple interface allows setting various registration settings.
Of note, we read in a parameter file (test2d_tst.json) to parametrize the registration algorithm and
write out the used parameters (after the run) into the same file as well as into a file with comments explaining
all the parameter settings. If params is not specified default settings will be used. By running for one
iteration for example, this allows writing out a fresh settings template which can then be edited.
si.register_images(I0, I1, spacing, model_name='svf_scalar_momentum_map',
nr_of_iterations=100,
use_multi_scale=False,
visualize_step=5,
optimizer_name='sgd',
learning_rate=0.1,
rel_ftol=1e-7,
json_config_out_filename=('test2d_tst.json',
'test2d_tst_with_comments.json'),
params='test2d_tst.json',
recording_step=None)
Out:
Loading parameter file = test2d_tst.json
Overwriting key = use_map; category = root.model.deformation; value = True -> True
Overwriting key = type; category = root.model.registration_model; value = svf_scalar_momentum_map -> svf_scalar_momentum_map
Overwriting key = name; category = root.optimizer; value = sgd -> sgd
Overwriting key = nr_of_iterations; category = root.optimizer.single_scale; value = 100 -> 100
Overwriting key = rel_ftol; category = root.optimizer.single_scale; value = 1e-07 -> 1e-07
Overwriting key = use_map; category = root.model.deformation; value = True -> True
Overwriting key = map_low_res_factor; category = root.model.deformation; value = None -> None
Overwriting key = type; category = root.model.registration_model; value = svf_scalar_momentum_map -> svf_scalar_momentum_map
Overwriting key = type; category = root.model.registration_model; value = svf_scalar_momentum_map -> svf_scalar_momentum_map
Using map-based svf_scalar_momentum_map model
works in mermaid iter mode
SVFScalarMomentumMapNet(
(integrator): ODEWrapBlock()
)
Overwriting key = lr; category = root.optimizer.sgd.individual; value = 0.1 -> 0.1
Overwriting key = lr; category = root.optimizer.sgd.shared; value = 0.1 -> 0.1
Using ncc similarity measure
[31m 0-Tot: E=069.5632 | simE=069.5632 | regE=000.0000 | optParE=000.0000 | relF= n/a | [0m
[34m 0-Img: E=069.5632 | simE=069.5632 | regE=000.0000 |[0m
[31m 1-Tot: E=061.8676 | simE=061.8671 | regE=000.0005 | optParE=000.0000 | relF=000.1224 | [0m
[34m 1-Img: E=061.8676 | simE=061.8671 | regE=000.0005 |[0m
[31m 2-Tot: E=049.0070 | simE=049.0051 | regE=000.0019 | optParE=000.0000 | relF=000.2572 | [0m
[34m 2-Img: E=049.0070 | simE=049.0051 | regE=000.0019 |[0m
[31m 3-Tot: E=033.5135 | simE=033.5080 | regE=000.0055 | optParE=000.0000 | relF=000.4489 | [0m
[34m 3-Img: E=033.5135 | simE=033.5080 | regE=000.0055 |[0m
[31m 4-Tot: E=020.3635 | simE=020.3540 | regE=000.0095 | optParE=000.0000 | relF=000.6155 | [0m
[34m 4-Img: E=020.3635 | simE=020.3540 | regE=000.0095 |[0m
[31m 5-Tot: E=016.4024 | simE=016.3870 | regE=000.0154 | optParE=000.0000 | relF=000.2276 | [0m
[34m 5-Img: E=016.4024 | simE=016.3870 | regE=000.0154 |[0m
[31m 6-Tot: E=019.1888 | simE=019.1709 | regE=000.0180 | optParE=000.0000 | relF=000.1380 | [0m
[34m 6-Img: E=019.1888 | simE=019.1709 | regE=000.0180 |[0m
[31m 7-Tot: E=018.8280 | simE=018.8096 | regE=000.0184 | optParE=000.0000 | relF=000.0182 | [0m
[34m 7-Img: E=018.8280 | simE=018.8096 | regE=000.0184 |[0m
[31m 8-Tot: E=020.8841 | simE=020.8677 | regE=000.0164 | optParE=000.0000 | relF=000.0940 | [0m
[34m 8-Img: E=020.8841 | simE=020.8677 | regE=000.0164 |[0m
[31m 9-Tot: E=015.0106 | simE=014.9945 | regE=000.0161 | optParE=000.0000 | relF=000.3668 | [0m
[34m 9-Img: E=015.0106 | simE=014.9945 | regE=000.0161 |[0m
[31m 10-Tot: E=011.6601 | simE=011.6479 | regE=000.0121 | optParE=000.0000 | relF=000.2647 | [0m
[34m 10-Img: E=011.6601 | simE=011.6479 | regE=000.0121 |[0m
[31m 11-Tot: E=010.9093 | simE=010.8990 | regE=000.0103 | optParE=000.0000 | relF=000.0630 | [0m
[34m 11-Img: E=010.9093 | simE=010.8990 | regE=000.0103 |[0m
[31m 12-Tot: E=011.3919 | simE=011.3812 | regE=000.0107 | optParE=000.0000 | relF=000.0389 | [0m
[34m 12-Img: E=011.3919 | simE=011.3812 | regE=000.0107 |[0m
[31m 13-Tot: E=010.1348 | simE=010.1239 | regE=000.0109 | optParE=000.0000 | relF=000.1129 | [0m
[34m 13-Img: E=010.1348 | simE=010.1239 | regE=000.0109 |[0m
[31m 14-Tot: E=009.7908 | simE=009.7788 | regE=000.0121 | optParE=000.0000 | relF=000.0319 | [0m
[34m 14-Img: E=009.7908 | simE=009.7788 | regE=000.0121 |[0m
[31m 15-Tot: E=009.8653 | simE=009.8504 | regE=000.0149 | optParE=000.0000 | relF=000.0069 | [0m
[34m 15-Img: E=009.8653 | simE=009.8504 | regE=000.0149 |[0m
[31m 16-Tot: E=011.9485 | simE=011.9354 | regE=000.0130 | optParE=000.0000 | relF=000.1609 | [0m
[34m 16-Img: E=011.9485 | simE=011.9354 | regE=000.0130 |[0m
[31m 17-Tot: E=006.7032 | simE=006.6900 | regE=000.0132 | optParE=000.0000 | relF=000.6809 | [0m
[34m 17-Img: E=006.7032 | simE=006.6900 | regE=000.0132 |[0m
[31m 18-Tot: E=012.9592 | simE=012.9464 | regE=000.0128 | optParE=000.0000 | relF=000.4482 | [0m
[34m 18-Img: E=012.9592 | simE=012.9464 | regE=000.0128 |[0m
[31m 19-Tot: E=011.0977 | simE=011.0852 | regE=000.0124 | optParE=000.0000 | relF=000.1539 | [0m
[34m 19-Img: E=011.0977 | simE=011.0852 | regE=000.0124 |[0m
[31m 20-Tot: E=005.8190 | simE=005.8059 | regE=000.0131 | optParE=000.0000 | relF=000.7741 | [0m
[34m 20-Img: E=005.8190 | simE=005.8059 | regE=000.0131 |[0m
[31m 21-Tot: E=010.2607 | simE=010.2444 | regE=000.0163 | optParE=000.0000 | relF=000.3944 | [0m
[34m 21-Img: E=010.2607 | simE=010.2444 | regE=000.0163 |[0m
[31m 22-Tot: E=013.9306 | simE=013.9116 | regE=000.0191 | optParE=000.0000 | relF=000.2458 | [0m
[34m 22-Img: E=013.9306 | simE=013.9116 | regE=000.0191 |[0m
[31m 23-Tot: E=010.2365 | simE=010.2196 | regE=000.0169 | optParE=000.0000 | relF=000.3288 | [0m
[34m 23-Img: E=010.2365 | simE=010.2196 | regE=000.0169 |[0m
[31m 24-Tot: E=007.9760 | simE=007.9627 | regE=000.0133 | optParE=000.0000 | relF=000.2518 | [0m
[34m 24-Img: E=007.9760 | simE=007.9627 | regE=000.0133 |[0m
[31m 25-Tot: E=009.7528 | simE=009.7396 | regE=000.0132 | optParE=000.0000 | relF=000.1652 | [0m
[34m 25-Img: E=009.7528 | simE=009.7396 | regE=000.0132 |[0m
[31m 26-Tot: E=008.4217 | simE=008.4079 | regE=000.0138 | optParE=000.0000 | relF=000.1413 | [0m
[34m 26-Img: E=008.4217 | simE=008.4079 | regE=000.0138 |[0m
[31m 27-Tot: E=007.8224 | simE=007.8081 | regE=000.0142 | optParE=000.0000 | relF=000.0679 | [0m
[34m 27-Img: E=007.8224 | simE=007.8081 | regE=000.0142 |[0m
[31m 28-Tot: E=008.0470 | simE=008.0320 | regE=000.0150 | optParE=000.0000 | relF=000.0248 | [0m
[34m 28-Img: E=008.0470 | simE=008.0320 | regE=000.0150 |[0m
[31m 29-Tot: E=006.4322 | simE=006.4181 | regE=000.0141 | optParE=000.0000 | relF=000.2173 | [0m
[34m 29-Img: E=006.4322 | simE=006.4181 | regE=000.0141 |[0m
[31m 30-Tot: E=009.5180 | simE=009.5023 | regE=000.0157 | optParE=000.0000 | relF=000.2934 | [0m
[34m 30-Img: E=009.5180 | simE=009.5023 | regE=000.0157 |[0m
Epoch 31: reducing learning rate of group 0 to 5.0000e-02.
Epoch 31: reducing learning rate of group 1 to 5.0000e-02.
[31m 31-Tot: E=007.3418 | simE=007.3264 | regE=000.0154 | optParE=000.0000 | relF=000.2609 | [0m
[34m 31-Img: E=007.3418 | simE=007.3264 | regE=000.0154 |[0m
[31m 32-Tot: E=007.4306 | simE=007.4165 | regE=000.0140 | optParE=000.0000 | relF=000.0105 | [0m
[34m 32-Img: E=007.4306 | simE=007.4165 | regE=000.0140 |[0m
[31m 33-Tot: E=004.4076 | simE=004.3934 | regE=000.0142 | optParE=000.0000 | relF=000.5590 | [0m
[34m 33-Img: E=004.4076 | simE=004.3934 | regE=000.0142 |[0m
[31m 34-Tot: E=006.0336 | simE=006.0179 | regE=000.0157 | optParE=000.0000 | relF=000.2312 | [0m
[34m 34-Img: E=006.0336 | simE=006.0179 | regE=000.0157 |[0m
[31m 35-Tot: E=004.9244 | simE=004.9091 | regE=000.0153 | optParE=000.0000 | relF=000.1872 | [0m
[34m 35-Img: E=004.9244 | simE=004.9091 | regE=000.0153 |[0m
[31m 36-Tot: E=004.5009 | simE=004.4870 | regE=000.0139 | optParE=000.0000 | relF=000.0770 | [0m
[34m 36-Img: E=004.5009 | simE=004.4870 | regE=000.0139 |[0m
[31m 37-Tot: E=004.3886 | simE=004.3739 | regE=000.0147 | optParE=000.0000 | relF=000.0208 | [0m
[34m 37-Img: E=004.3886 | simE=004.3739 | regE=000.0147 |[0m
[31m 38-Tot: E=005.3785 | simE=005.3642 | regE=000.0143 | optParE=000.0000 | relF=000.1552 | [0m
[34m 38-Img: E=005.3785 | simE=005.3642 | regE=000.0143 |[0m
[31m 39-Tot: E=004.0750 | simE=004.0610 | regE=000.0139 | optParE=000.0000 | relF=000.2569 | [0m
[34m 39-Img: E=004.0750 | simE=004.0610 | regE=000.0139 |[0m
[31m 40-Tot: E=004.8721 | simE=004.8574 | regE=000.0147 | optParE=000.0000 | relF=000.1357 | [0m
[34m 40-Img: E=004.8721 | simE=004.8574 | regE=000.0147 |[0m
[31m 41-Tot: E=004.1322 | simE=004.1179 | regE=000.0143 | optParE=000.0000 | relF=000.1442 | [0m
[34m 41-Img: E=004.1322 | simE=004.1179 | regE=000.0143 |[0m
[31m 42-Tot: E=004.5222 | simE=004.5082 | regE=000.0140 | optParE=000.0000 | relF=000.0706 | [0m
[34m 42-Img: E=004.5222 | simE=004.5082 | regE=000.0140 |[0m
[31m 43-Tot: E=004.0501 | simE=004.0359 | regE=000.0141 | optParE=000.0000 | relF=000.0935 | [0m
[34m 43-Img: E=004.0501 | simE=004.0359 | regE=000.0141 |[0m
[31m 44-Tot: E=004.3232 | simE=004.3091 | regE=000.0141 | optParE=000.0000 | relF=000.0513 | [0m
[34m 44-Img: E=004.3232 | simE=004.3091 | regE=000.0141 |[0m
[31m 45-Tot: E=004.0164 | simE=004.0025 | regE=000.0140 | optParE=000.0000 | relF=000.0612 | [0m
[34m 45-Img: E=004.0164 | simE=004.0025 | regE=000.0140 |[0m
[31m 46-Tot: E=004.1650 | simE=004.1508 | regE=000.0142 | optParE=000.0000 | relF=000.0288 | [0m
[34m 46-Img: E=004.1650 | simE=004.1508 | regE=000.0142 |[0m
[31m 47-Tot: E=003.9848 | simE=003.9710 | regE=000.0138 | optParE=000.0000 | relF=000.0361 | [0m
[34m 47-Img: E=003.9848 | simE=003.9710 | regE=000.0138 |[0m
[31m 48-Tot: E=004.0452 | simE=004.0309 | regE=000.0143 | optParE=000.0000 | relF=000.0120 | [0m
[34m 48-Img: E=004.0452 | simE=004.0309 | regE=000.0143 |[0m
[31m 49-Tot: E=003.9392 | simE=003.9253 | regE=000.0138 | optParE=000.0000 | relF=000.0215 | [0m
[34m 49-Img: E=003.9392 | simE=003.9253 | regE=000.0138 |[0m
[31m 50-Tot: E=003.9422 | simE=003.9279 | regE=000.0143 | optParE=000.0000 | relF=000.0006 | [0m
[34m 50-Img: E=003.9422 | simE=003.9279 | regE=000.0143 |[0m
[31m 51-Tot: E=003.9016 | simE=003.8877 | regE=000.0139 | optParE=000.0000 | relF=000.0083 | [0m
[34m 51-Img: E=003.9016 | simE=003.8877 | regE=000.0139 |[0m
[31m 52-Tot: E=003.8440 | simE=003.8296 | regE=000.0144 | optParE=000.0000 | relF=000.0119 | [0m
[34m 52-Img: E=003.8440 | simE=003.8296 | regE=000.0144 |[0m
[31m 53-Tot: E=003.8649 | simE=003.8510 | regE=000.0139 | optParE=000.0000 | relF=000.0043 | [0m
[34m 53-Img: E=003.8649 | simE=003.8510 | regE=000.0139 |[0m
[31m 54-Tot: E=003.7638 | simE=003.7494 | regE=000.0145 | optParE=000.0000 | relF=000.0212 | [0m
[34m 54-Img: E=003.7638 | simE=003.7494 | regE=000.0145 |[0m
[31m 55-Tot: E=003.8331 | simE=003.8191 | regE=000.0139 | optParE=000.0000 | relF=000.0143 | [0m
[34m 55-Img: E=003.8331 | simE=003.8191 | regE=000.0139 |[0m
[31m 56-Tot: E=003.7097 | simE=003.6952 | regE=000.0145 | optParE=000.0000 | relF=000.0262 | [0m
[34m 56-Img: E=003.7097 | simE=003.6952 | regE=000.0145 |[0m
[31m 57-Tot: E=003.8403 | simE=003.8262 | regE=000.0141 | optParE=000.0000 | relF=000.0270 | [0m
[34m 57-Img: E=003.8403 | simE=003.8262 | regE=000.0141 |[0m
[31m 58-Tot: E=003.8193 | simE=003.8048 | regE=000.0145 | optParE=000.0000 | relF=000.0044 | [0m
[34m 58-Img: E=003.8193 | simE=003.8048 | regE=000.0145 |[0m
[31m 59-Tot: E=004.0708 | simE=004.0563 | regE=000.0145 | optParE=000.0000 | relF=000.0496 | [0m
[34m 59-Img: E=004.0708 | simE=004.0563 | regE=000.0145 |[0m
[31m 60-Tot: E=003.5676 | simE=003.5531 | regE=000.0146 | optParE=000.0000 | relF=000.1102 | [0m
[34m 60-Img: E=003.5676 | simE=003.5531 | regE=000.0146 |[0m
[31m 61-Tot: E=004.1002 | simE=004.0857 | regE=000.0145 | optParE=000.0000 | relF=000.1044 | [0m
[34m 61-Img: E=004.1002 | simE=004.0857 | regE=000.0145 |[0m
[31m 62-Tot: E=003.3804 | simE=003.3660 | regE=000.0144 | optParE=000.0000 | relF=000.1643 | [0m
[34m 62-Img: E=003.3804 | simE=003.3660 | regE=000.0144 |[0m
[31m 63-Tot: E=004.1078 | simE=004.0933 | regE=000.0145 | optParE=000.0000 | relF=000.1424 | [0m
[34m 63-Img: E=004.1078 | simE=004.0933 | regE=000.0145 |[0m
[31m 64-Tot: E=003.2580 | simE=003.2437 | regE=000.0143 | optParE=000.0000 | relF=000.1996 | [0m
[34m 64-Img: E=003.2580 | simE=003.2437 | regE=000.0143 |[0m
[31m 65-Tot: E=004.1951 | simE=004.1804 | regE=000.0148 | optParE=000.0000 | relF=000.1804 | [0m
[34m 65-Img: E=004.1951 | simE=004.1804 | regE=000.0148 |[0m
[31m 66-Tot: E=003.1704 | simE=003.1562 | regE=000.0142 | optParE=000.0000 | relF=000.2457 | [0m
[34m 66-Img: E=003.1704 | simE=003.1562 | regE=000.0142 |[0m
[31m 67-Tot: E=004.2748 | simE=004.2599 | regE=000.0149 | optParE=000.0000 | relF=000.2094 | [0m
[34m 67-Img: E=004.2748 | simE=004.2599 | regE=000.0149 |[0m
[31m 68-Tot: E=003.1301 | simE=003.1158 | regE=000.0143 | optParE=000.0000 | relF=000.2772 | [0m
[34m 68-Img: E=003.1301 | simE=003.1158 | regE=000.0143 |[0m
[31m 69-Tot: E=004.1197 | simE=004.1052 | regE=000.0145 | optParE=000.0000 | relF=000.1933 | [0m
[34m 69-Img: E=004.1197 | simE=004.1052 | regE=000.0145 |[0m
[31m 70-Tot: E=003.0977 | simE=003.0834 | regE=000.0143 | optParE=000.0000 | relF=000.2494 | [0m
[34m 70-Img: E=003.0977 | simE=003.0834 | regE=000.0143 |[0m
[31m 71-Tot: E=004.2609 | simE=004.2459 | regE=000.0150 | optParE=000.0000 | relF=000.2211 | [0m
[34m 71-Img: E=004.2609 | simE=004.2459 | regE=000.0150 |[0m
[31m 72-Tot: E=003.4856 | simE=003.4710 | regE=000.0146 | optParE=000.0000 | relF=000.1728 | [0m
[34m 72-Img: E=003.4856 | simE=003.4710 | regE=000.0146 |[0m
[31m 73-Tot: E=003.6351 | simE=003.6213 | regE=000.0139 | optParE=000.0000 | relF=000.0323 | [0m
[34m 73-Img: E=003.6351 | simE=003.6213 | regE=000.0139 |[0m
[31m 74-Tot: E=003.0905 | simE=003.0760 | regE=000.0145 | optParE=000.0000 | relF=000.1331 | [0m
[34m 74-Img: E=003.0905 | simE=003.0760 | regE=000.0145 |[0m
[31m 75-Tot: E=004.0408 | simE=004.0266 | regE=000.0142 | optParE=000.0000 | relF=000.1885 | [0m
[34m 75-Img: E=004.0408 | simE=004.0266 | regE=000.0142 |[0m
[31m 76-Tot: E=002.9283 | simE=002.9137 | regE=000.0146 | optParE=000.0000 | relF=000.2832 | [0m
[34m 76-Img: E=002.9283 | simE=002.9137 | regE=000.0146 |[0m
[31m 77-Tot: E=004.5524 | simE=004.5381 | regE=000.0143 | optParE=000.0000 | relF=000.2925 | [0m
[34m 77-Img: E=004.5524 | simE=004.5381 | regE=000.0143 |[0m
[31m 78-Tot: E=002.8665 | simE=002.8519 | regE=000.0145 | optParE=000.0000 | relF=000.4361 | [0m
[34m 78-Img: E=002.8665 | simE=002.8519 | regE=000.0145 |[0m
[31m 79-Tot: E=005.2399 | simE=005.2238 | regE=000.0162 | optParE=000.0000 | relF=000.3804 | [0m
[34m 79-Img: E=005.2399 | simE=005.2238 | regE=000.0162 |[0m
[31m 80-Tot: E=003.0751 | simE=003.0597 | regE=000.0154 | optParE=000.0000 | relF=000.5312 | [0m
[34m 80-Img: E=003.0751 | simE=003.0597 | regE=000.0154 |[0m
[31m 81-Tot: E=003.8405 | simE=003.8267 | regE=000.0138 | optParE=000.0000 | relF=000.1581 | [0m
[34m 81-Img: E=003.8405 | simE=003.8267 | regE=000.0138 |[0m
[31m 82-Tot: E=003.1878 | simE=003.1734 | regE=000.0145 | optParE=000.0000 | relF=000.1559 | [0m
[34m 82-Img: E=003.1878 | simE=003.1734 | regE=000.0145 |[0m
[31m 83-Tot: E=003.7209 | simE=003.7055 | regE=000.0154 | optParE=000.0000 | relF=000.1129 | [0m
[34m 83-Img: E=003.7209 | simE=003.7055 | regE=000.0154 |[0m
[31m 84-Tot: E=002.9631 | simE=002.9483 | regE=000.0149 | optParE=000.0000 | relF=000.1912 | [0m
[34m 84-Img: E=002.9631 | simE=002.9483 | regE=000.0149 |[0m
[31m 85-Tot: E=003.9692 | simE=003.9540 | regE=000.0152 | optParE=000.0000 | relF=000.2025 | [0m
[34m 85-Img: E=003.9692 | simE=003.9540 | regE=000.0152 |[0m
[31m 86-Tot: E=002.8231 | simE=002.8084 | regE=000.0148 | optParE=000.0000 | relF=000.2998 | [0m
[34m 86-Img: E=002.8231 | simE=002.8084 | regE=000.0148 |[0m
[31m 87-Tot: E=004.2305 | simE=004.2151 | regE=000.0154 | optParE=000.0000 | relF=000.2691 | [0m
[34m 87-Img: E=004.2305 | simE=004.2151 | regE=000.0154 |[0m
[31m 88-Tot: E=002.7439 | simE=002.7291 | regE=000.0148 | optParE=000.0000 | relF=000.3971 | [0m
[34m 88-Img: E=002.7439 | simE=002.7291 | regE=000.0148 |[0m
[31m 89-Tot: E=004.3772 | simE=004.3619 | regE=000.0153 | optParE=000.0000 | relF=000.3038 | [0m
[34m 89-Img: E=004.3772 | simE=004.3619 | regE=000.0153 |[0m
[31m 90-Tot: E=002.6952 | simE=002.6803 | regE=000.0149 | optParE=000.0000 | relF=000.4552 | [0m
[34m 90-Img: E=002.6952 | simE=002.6803 | regE=000.0149 |[0m
[31m 91-Tot: E=004.4210 | simE=004.4059 | regE=000.0151 | optParE=000.0000 | relF=000.3184 | [0m
[34m 91-Img: E=004.4210 | simE=004.4059 | regE=000.0151 |[0m
[31m 92-Tot: E=002.6718 | simE=002.6570 | regE=000.0147 | optParE=000.0000 | relF=000.4764 | [0m
[34m 92-Img: E=002.6718 | simE=002.6570 | regE=000.0147 |[0m
[31m 93-Tot: E=004.0538 | simE=004.0380 | regE=000.0158 | optParE=000.0000 | relF=000.2735 | [0m
[34m 93-Img: E=004.0538 | simE=004.0380 | regE=000.0158 |[0m
[31m 94-Tot: E=002.6442 | simE=002.6291 | regE=000.0151 | optParE=000.0000 | relF=000.3868 | [0m
[34m 94-Img: E=002.6442 | simE=002.6291 | regE=000.0151 |[0m
[31m 95-Tot: E=004.1189 | simE=004.1037 | regE=000.0152 | optParE=000.0000 | relF=000.2881 | [0m
[34m 95-Img: E=004.1189 | simE=004.1037 | regE=000.0152 |[0m
[31m 96-Tot: E=002.7061 | simE=002.6914 | regE=000.0148 | optParE=000.0000 | relF=000.3812 | [0m
[34m 96-Img: E=002.7061 | simE=002.6914 | regE=000.0148 |[0m
[31m 97-Tot: E=004.1238 | simE=004.1080 | regE=000.0158 | optParE=000.0000 | relF=000.2767 | [0m
[34m 97-Img: E=004.1238 | simE=004.1080 | regE=000.0158 |[0m
[31m 98-Tot: E=002.6161 | simE=002.6009 | regE=000.0152 | optParE=000.0000 | relF=000.4169 | [0m
[34m 98-Img: E=002.6161 | simE=002.6009 | regE=000.0152 |[0m
[31m 99-Tot: E=004.3099 | simE=004.2947 | regE=000.0151 | optParE=000.0000 | relF=000.3190 | [0m
[34m 99-Img: E=004.3099 | simE=004.2947 | regE=000.0151 |[0m
[32m-->Elapsed time 43.61767[s][0m
Writing parameter file = test2d_tst.json
Writing parameter file = test2d_tst_with_comments.json
Plotting some results¶
We can query the energies over the iterations. Note that this code need to be modified for a multi-scale solution as energies will be returned at each scale.
h = si.get_history()
plt.clf()
e_p, = plt.plot(h['energy'], label='energy')
s_p, = plt.plot(h['similarity_energy'], label='similarity_energy')
r_p, = plt.plot(h['regularization_energy'], label='regularization_energy')
plt.legend(handles=[e_p, s_p, r_p])
plt.show()
Total running time of the script: ( 0 minutes 44.136 seconds)