def clm_build_benchmark(training_images, training_options=None, verbose=False):
r"""
Builds an CLM model.
Parameters
----------
training_images: list of :class:MaskedImage objects
A list of the training images.
training_options: dictionary, optional
A dictionary with the parameters that will be passed in the CLMBuilder
(:class:menpo.fitmultilevel.clm.CLMBuilder).
If None, the default options will be used.
This is an example of the dictionary with the default options:
training_options = {'group': 'PTS',
'classifier_type': linear_svm_lr,
'patch_shape': (5, 5),
'feature_type': sparse_hog,
'normalization_diagonal': None,
'n_levels': 3,
'downscale': 1.1,
'scaled_shape_models': True,
'pyramid_on_features': True,
'max_shape_components': None,
'boundary': 3,
'interpolator': 'scipy'
}
For an explanation of the options, please refer to the CLMBuilder
documentation.
Default: None
verbose: boolean, optional
If True, it prints information regarding the CLM training.
Default: False
Returns
-------
clm: :class:menpo.fitmultilevel.clm.CLM object
The trained CLM model.
"""
if verbose:
print('CLM Training:')
# parse options
if training_options is None:
training_options = {}
# group option
group = training_options.pop('group', None)
# build aam
aam = CLMBuilder(**training_options).build(training_images,
group=group,
verbose=verbose)
return aam
regression_features=weights,
noise_std=0.04,
rotation=False,
n_perturbations=1,
update='compositional',
md_transform=OrthoMDTransform,
global_transform=AlignmentSimilarity,
n_shape=25,
n_appearance=None).train(training_images, group='PTS')
clm = CLMBuilder(classifier_type=linear_svm_lr,
feature_type=[sparse_hog],
normalization_diagonal=100,
patch_shape=(5, 5),
n_levels=1,
downscale=1.1,
scaled_shape_models=True,
pyramid_on_features=True,
max_shape_components=25,
boundary=3,
interpolator='scipy').build(training_images, group='PTS')
sdm3 = SDCLMTrainer(clm,
regression_type=mlr,
noise_std=0.04,
rotation=False,
n_perturbations=1,
pdm_transform=OrthoPDM,
global_transform=AlignmentSimilarity,
n_shape=None).train(training_images, group='PTS')
def test_feature_type_with_pyramid_on_features_exception():
clm = CLMBuilder(feature_type=['igo', sparse_hog]).build(training_images,
group='PTS')
def test_classifier_type_2_exception():
clm = CLMBuilder(classifier_type=['linear_svm_lr']).build(
training_images, group='PTS')
def test_patch_shape_2_exception():
clm = CLMBuilder(patch_shape=(5, 6, 7)).build(training_images, group='PTS')
def test_boundary_exception():
clm = CLMBuilder(boundary=-1).build(training_images, group='PTS')
def test_verbose_mock(mock_stdout):
clm = CLMBuilder().build(training_images, group='PTS', verbose=True)
def test_max_shape_components_2_exception():
clm = CLMBuilder(max_shape_components=[1, 2]).build(training_images,
group='PTS')
def test_normalization_diagonal_exception():
clm = CLMBuilder(normalization_diagonal=10).build(training_images,
group='PTS')
def test_downscale_exception():
clm = CLMBuilder(downscale=1).build(training_images,
group='PTS')
assert (clm.downscale == 1)
clm = CLMBuilder(downscale=0).build(training_images,
group='PTS')
def test_n_levels_exception():
clm = CLMBuilder(n_levels=0).build(training_images,
group='PTS')
training_images = []
for i in range(4):
im = mio.import_builtin_asset(filenames[i])
im.crop_to_landmarks_proportion_inplace(0.1)
labeller(im, 'PTS', ibug_68_trimesh)
if im.n_channels == 3:
im = im.as_greyscale(mode='luminosity')
training_images.append(im)
# build clms
clm1 = CLMBuilder(classifier_type=[linear_svm_lr],
patch_shape=(5, 5),
feature_type=['igo', sparse_hog, None],
normalization_diagonal=150,
n_levels=3,
downscale=2,
scaled_shape_models=False,
pyramid_on_features=False,
max_shape_components=[1, 2, 3],
boundary=3,
interpolator='scipy').build(training_images, group='PTS')
clm2 = CLMBuilder(classifier_type=[random_forest, linear_svm_lr],
patch_shape=(3, 10),
feature_type=None,
normalization_diagonal=None,
n_levels=2,
downscale=1.2,
scaled_shape_models=True,
pyramid_on_features=False,
max_shape_components=None,