torchio.transforms.augmentation.composition.Compose.__init__() - Code Metrics - Inspection of "Fix flake8 errors" - fepegar/torchio - Measure and Improve Code Quality continuously with Scrutinizer

Passed

Push — master ( db49f2...b2e640 )

by Fernando

created 2020-04-20 15:51 UTC

Compose.init() A

↳ Parent: torchio.transforms.augmentation.composition

Complexity

Conditions

Size

Total Lines	3
Code Lines	3

Duplication

Lines	0
Ratio	0 %

Importance

Changes

Metric	Value
cc	1
eloc	3
nop	3
dl	0
loc	3
rs	10
c	0
b	0
f	0

from typing import Union, Sequence

import torch

import numpy as np

from torchvision.transforms import Compose as PyTorchCompose


from .. import Transform
from . import RandomTransform


class Compose(Transform):
    """Compose several transforms together.

    Args:
        transforms: list of instances of
            :py:class:`~torchio.transforms.transform.Transform`.
        p: Probability that this transform will be applied.

    .. note::
        This is a thin wrapper of :py:class:`torchvision.transforms.Compose`.
    """
    def __init__(self, transforms: Sequence[Transform], p: float = 1):
        super().__init__(p=p)
        self.transform = PyTorchCompose(transforms)

    def apply_transform(self, sample: dict):
        return self.transform(sample)


class OneOf(RandomTransform):
    """Apply only one of the given transforms.

    Args:
        transforms: Dictionary with instances of
            :py:class:`~torchio.transforms.transform.Transform` as keys and
            probabilities as values. Probabilities are normalized so they sum
            to one. If a sequence is given, the same probability will be
            assigned to each transform.
        p: Probability that this transform will be applied.

    Example:
        >>> import torchio
        >>> ixi = torchio.datasets.ixi.IXITiny('ixi', download=True)
        >>> sample = ixi[0]
        >>> transforms_dict = {
        ...     torchio.transforms.RandomAffine(): 0.75,
        ...     torchio.transforms.RandomElasticDeformation(): 0.25,
        ... }  # Using 3 and 1 as probabilities would have the same effect
        >>> transform = torchio.transforms.OneOf(transforms_dict)

    """
    def __init__(self, transforms: Union[dict, Sequence], p: float = 1):
        super().__init__(p=p)
        self.transforms_dict = self._get_transforms_dict(transforms)

    def apply_transform(self, sample: dict):
        weights = torch.Tensor(list(self.transforms_dict.values()))
        index = torch.multinomial(weights, 1)
        transforms = list(self.transforms_dict.keys())
        transform = transforms[index]
        transformed = transform(sample)
        return transformed

    def _get_transforms_dict(self, transforms: Union[dict, Sequence]):
        if isinstance(transforms, dict):
            transforms_dict = dict(transforms)
            self._normalize_probabilities(transforms_dict)
        else:
            try:
                p = 1 / len(transforms)

            except TypeError as e:

                message = (
                    'Transforms argument must be a dictionary or a sequence,'
                    f' not {type(transforms)}'
                )
                raise ValueError(message) from e
            transforms_dict = {transform: p for transform in transforms}
        for transform in transforms_dict:
            if not isinstance(transform, Transform):
                message = (
                    'All keys in transform_dict must be instances of'
                    f'torchio.Transform, not "{type(transform)}"'
                )
                raise ValueError(message)
        return transforms_dict

    @staticmethod
    def _normalize_probabilities(transforms_dict: dict):
        probabilities = np.array(list(transforms_dict.values()), dtype=float)
        if np.any(probabilities < 0):
            message = (
                'Probabilities must be greater or equal to zero,'
                f' not "{probabilities}"'
            )
            raise ValueError(message)
        if np.all(probabilities == 0):
            message = (
                'At least one probability must be greater than zero,'
                f' but they are "{probabilities}"'
            )
            raise ValueError(message)
        for transform, probability in transforms_dict.items():
            transforms_dict[transform] = probability / probabilities.sum()


1			from typing import Union, Sequence
2
3			import torch
			0 ignored issues – show introduced 2020-04-20 15:52 UTC by Report Bug Copy Issue Report Unable to import 'torch' Loading history...
4			import numpy as np
			0 ignored issues – show introduced 2020-04-20 15:52 UTC by Report Bug Copy Issue Report Unable to import 'numpy' Loading history...
5			from torchvision.transforms import Compose as PyTorchCompose
			0 ignored issues – show introduced 2020-04-20 15:52 UTC by Report Bug Copy Issue Report Unable to import 'torchvision.transforms' Loading history...
6
7			from .. import Transform
8			from . import RandomTransform
9
10
11			class Compose(Transform):
12			"""Compose several transforms together.
13
14			Args:
15			transforms: list of instances of
16			:py:class:`~torchio.transforms.transform.Transform`.
17			p: Probability that this transform will be applied.
18
19			.. note::
20			This is a thin wrapper of :py:class:`torchvision.transforms.Compose`.
21			"""
22			def __init__(self, transforms: Sequence[Transform], p: float = 1):
23			super().__init__(p=p)
24			self.transform = PyTorchCompose(transforms)
25
26			def apply_transform(self, sample: dict):
27			return self.transform(sample)
28
29
30			class OneOf(RandomTransform):
31			"""Apply only one of the given transforms.
32
33			Args:
34			transforms: Dictionary with instances of
35			:py:class:`~torchio.transforms.transform.Transform` as keys and
36			probabilities as values. Probabilities are normalized so they sum
37			to one. If a sequence is given, the same probability will be
38			assigned to each transform.
39			p: Probability that this transform will be applied.
40
41			Example:
42			>>> import torchio
43			>>> ixi = torchio.datasets.ixi.IXITiny('ixi', download=True)
44			>>> sample = ixi[0]
45			>>> transforms_dict = {
46			... torchio.transforms.RandomAffine(): 0.75,
47			... torchio.transforms.RandomElasticDeformation(): 0.25,
48			... } # Using 3 and 1 as probabilities would have the same effect
49			>>> transform = torchio.transforms.OneOf(transforms_dict)
50
51			"""
52			def __init__(self, transforms: Union[dict, Sequence], p: float = 1):
53			super().__init__(p=p)
54			self.transforms_dict = self._get_transforms_dict(transforms)
55
56			def apply_transform(self, sample: dict):
57			weights = torch.Tensor(list(self.transforms_dict.values()))
58			index = torch.multinomial(weights, 1)
59			transforms = list(self.transforms_dict.keys())
60			transform = transforms[index]
61			transformed = transform(sample)
62			return transformed
63
64			def _get_transforms_dict(self, transforms: Union[dict, Sequence]):
65			if isinstance(transforms, dict):
66			transforms_dict = dict(transforms)
67			self._normalize_probabilities(transforms_dict)
68			else:
69			try:
70			p = 1 / len(transforms)
			0 ignored issues – show Coding Style Naming introduced 2020-04-20 15:52 UTC by Report Bug Copy Issue Report Variable name "p" doesn't conform to snake_case naming style ('(([a-z_][a-z0-9_]2,)\|(_[a-z0-9_]*)\|(__[a-z][a-z0-9_]+__))$' pattern) This check looks for invalid names for a range of different identifiers. You can set regular expressions to which the identifiers must conform if the defaults do not match your requirements. If your project includes a Pylint configuration file, the settings contained in that file take precedence. To find out more about Pylint, please refer to their site. Loading history...
71			except TypeError as e:
			0 ignored issues – show Coding Style Naming introduced 2020-04-20 15:52 UTC by Report Bug Copy Issue Report Variable name "e" doesn't conform to snake_case naming style ('(([a-z_][a-z0-9_]2,)\|(_[a-z0-9_]*)\|(__[a-z][a-z0-9_]+__))$' pattern) This check looks for invalid names for a range of different identifiers. You can set regular expressions to which the identifiers must conform if the defaults do not match your requirements. If your project includes a Pylint configuration file, the settings contained in that file take precedence. To find out more about Pylint, please refer to their site. Loading history...
72			message = (
73			'Transforms argument must be a dictionary or a sequence,'
74			f' not {type(transforms)}'
75			)
76			raise ValueError(message) from e
77			transforms_dict = {transform: p for transform in transforms}
78			for transform in transforms_dict:
79			if not isinstance(transform, Transform):
80			message = (
81			'All keys in transform_dict must be instances of'
82			f'torchio.Transform, not "{type(transform)}"'
83			)
84			raise ValueError(message)
85			return transforms_dict
86
87			@staticmethod
88			def _normalize_probabilities(transforms_dict: dict):
89			probabilities = np.array(list(transforms_dict.values()), dtype=float)
90			if np.any(probabilities < 0):
91			message = (
92			'Probabilities must be greater or equal to zero,'
93			f' not "{probabilities}"'
94			)
95			raise ValueError(message)
96			if np.all(probabilities == 0):
97			message = (
98			'At least one probability must be greater than zero,'
99			f' but they are "{probabilities}"'
100			)
101			raise ValueError(message)
102			for transform, probability in transforms_dict.items():
103			transforms_dict[transform] = probability / probabilities.sum()
104

fepegar / torchio

Push — master ( db49f2...b2e640 )

Compose.__init__() A

Complexity

Size

Duplication

Importance

Duplication Side-by-Side

Filter issues like

Compose.init() A