-
Notifications
You must be signed in to change notification settings - Fork 387
/
hinge.py
289 lines (238 loc) · 12.1 KB
/
hinge.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
# Copyright The Lightning team.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
from typing import Optional, Tuple
import torch
from torch import Tensor, tensor
from typing_extensions import Literal
from torchmetrics.functional.classification.confusion_matrix import (
_binary_confusion_matrix_format,
_binary_confusion_matrix_tensor_validation,
_multiclass_confusion_matrix_format,
_multiclass_confusion_matrix_tensor_validation,
)
from torchmetrics.utilities.data import to_onehot
from torchmetrics.utilities.enums import ClassificationTaskNoMultilabel
def _hinge_loss_compute(measure: Tensor, total: Tensor) -> Tensor:
return measure / total
def _binary_hinge_loss_arg_validation(squared: bool, ignore_index: Optional[int] = None) -> None:
if not isinstance(squared, bool):
raise ValueError(f"Expected argument `squared` to be an bool but got {squared}")
if ignore_index is not None and not isinstance(ignore_index, int):
raise ValueError(f"Expected argument `ignore_index` to either be `None` or an integer, but got {ignore_index}")
def _binary_hinge_loss_tensor_validation(preds: Tensor, target: Tensor, ignore_index: Optional[int] = None) -> None:
_binary_confusion_matrix_tensor_validation(preds, target, ignore_index)
if not preds.is_floating_point():
raise ValueError(
"Expected argument `preds` to be floating tensor with probabilities/logits"
f" but got tensor with dtype {preds.dtype}"
)
def _binary_hinge_loss_update(
preds: Tensor,
target: Tensor,
squared: bool,
) -> Tuple[Tensor, Tensor]:
target = target.bool()
margin = torch.zeros_like(preds)
margin[target] = preds[target]
margin[~target] = -preds[~target]
measures = 1 - margin
measures = torch.clamp(measures, 0)
if squared:
measures = measures.pow(2)
total = tensor(target.shape[0], device=target.device)
return measures.sum(dim=0), total
def binary_hinge_loss(
preds: Tensor,
target: Tensor,
squared: bool = False,
ignore_index: Optional[int] = None,
validate_args: bool = False,
) -> Tensor:
r"""Compute the mean `Hinge loss`_ typically used for Support Vector Machines (SVMs) for binary tasks.
.. math::
\text{Hinge loss} = \max(0, 1 - y \times \hat{y})
Where :math:`y \in {-1, 1}` is the target, and :math:`\hat{y} \in \mathbb{R}` is the prediction.
Accepts the following input tensors:
- ``preds`` (float tensor): ``(N, ...)``. Preds should be a tensor containing probabilities or logits for each
observation. If preds has values outside [0,1] range we consider the input to be logits and will auto apply
sigmoid per element.
- ``target`` (int tensor): ``(N, ...)``. Target should be a tensor containing ground truth labels, and therefore
only contain {0,1} values (except if `ignore_index` is specified). The value 1 always encodes the positive class.
Additional dimension ``...`` will be flattened into the batch dimension.
Args:
preds: Tensor with predictions
target: Tensor with true labels
squared:
If True, this will compute the squared hinge loss. Otherwise, computes the regular hinge loss.
ignore_index:
Specifies a target value that is ignored and does not contribute to the metric calculation
validate_args: bool indicating if input arguments and tensors should be validated for correctness.
Set to ``False`` for faster computations.
Example:
>>> from torch import tensor
>>> from torchmetrics.functional.classification import binary_hinge_loss
>>> preds = tensor([0.25, 0.25, 0.55, 0.75, 0.75])
>>> target = tensor([0, 0, 1, 1, 1])
>>> binary_hinge_loss(preds, target)
tensor(0.6900)
>>> binary_hinge_loss(preds, target, squared=True)
tensor(0.6905)
"""
if validate_args:
_binary_hinge_loss_arg_validation(squared, ignore_index)
_binary_hinge_loss_tensor_validation(preds, target, ignore_index)
preds, target = _binary_confusion_matrix_format(
preds, target, threshold=0.0, ignore_index=ignore_index, convert_to_labels=False
)
measures, total = _binary_hinge_loss_update(preds, target, squared)
return _hinge_loss_compute(measures, total)
def _multiclass_hinge_loss_arg_validation(
num_classes: int,
squared: bool = False,
multiclass_mode: Literal["crammer-singer", "one-vs-all"] = "crammer-singer",
ignore_index: Optional[int] = None,
) -> None:
_binary_hinge_loss_arg_validation(squared, ignore_index)
if not isinstance(num_classes, int) or num_classes < 2:
raise ValueError(f"Expected argument `num_classes` to be an integer larger than 1, but got {num_classes}")
allowed_mm = ("crammer-singer", "one-vs-all")
if multiclass_mode not in allowed_mm:
raise ValueError(f"Expected argument `multiclass_mode` to be one of {allowed_mm}, but got {multiclass_mode}.")
def _multiclass_hinge_loss_tensor_validation(
preds: Tensor, target: Tensor, num_classes: int, ignore_index: Optional[int] = None
) -> None:
_multiclass_confusion_matrix_tensor_validation(preds, target, num_classes, ignore_index)
if not preds.is_floating_point():
raise ValueError(
"Expected argument `preds` to be floating tensor with probabilities/logits"
f" but got tensor with dtype {preds.dtype}"
)
def _multiclass_hinge_loss_update(
preds: Tensor,
target: Tensor,
squared: bool,
multiclass_mode: Literal["crammer-singer", "one-vs-all"] = "crammer-singer",
) -> Tuple[Tensor, Tensor]:
if not torch.all((preds >= 0) * (preds <= 1)):
preds = preds.softmax(1)
target = to_onehot(target, max(2, preds.shape[1])).bool()
if multiclass_mode == "crammer-singer":
margin = preds[target]
margin -= torch.max(preds[~target].view(preds.shape[0], -1), dim=1)[0]
else:
target = target.bool()
margin = torch.zeros_like(preds)
margin[target] = preds[target]
margin[~target] = -preds[~target]
measures = 1 - margin
measures = torch.clamp(measures, 0)
if squared:
measures = measures.pow(2)
total = tensor(target.shape[0], device=target.device)
return measures.sum(dim=0), total
def multiclass_hinge_loss(
preds: Tensor,
target: Tensor,
num_classes: int,
squared: bool = False,
multiclass_mode: Literal["crammer-singer", "one-vs-all"] = "crammer-singer",
ignore_index: Optional[int] = None,
validate_args: bool = False,
) -> Tensor:
r"""Compute the mean `Hinge loss`_ typically used for Support Vector Machines (SVMs) for multiclass tasks.
The metric can be computed in two ways. Either, the definition by Crammer and Singer is used:
.. math::
\text{Hinge loss} = \max\left(0, 1 - \hat{y}_y + \max_{i \ne y} (\hat{y}_i)\right)
Where :math:`y \in {0, ..., \mathrm{C}}` is the target class (where :math:`\mathrm{C}` is the number of classes),
and :math:`\hat{y} \in \mathbb{R}^\mathrm{C}` is the predicted output per class. Alternatively, the metric can
also be computed in one-vs-all approach, where each class is valued against all other classes in a binary fashion.
Accepts the following input tensors:
- ``preds`` (float tensor): ``(N, C, ...)``. Preds should be a tensor containing probabilities or logits for each
observation. If preds has values outside [0,1] range we consider the input to be logits and will auto apply
softmax per sample.
- ``target`` (int tensor): ``(N, ...)``. Target should be a tensor containing ground truth labels, and therefore
only contain values in the [0, n_classes-1] range (except if `ignore_index` is specified).
Additional dimension ``...`` will be flattened into the batch dimension.
Args:
preds: Tensor with predictions
target: Tensor with true labels
num_classes: Integer specifing the number of classes
squared:
If True, this will compute the squared hinge loss. Otherwise, computes the regular hinge loss.
multiclass_mode:
Determines how to compute the metric
ignore_index:
Specifies a target value that is ignored and does not contribute to the metric calculation
validate_args: bool indicating if input arguments and tensors should be validated for correctness.
Set to ``False`` for faster computations.
Example:
>>> from torch import tensor
>>> from torchmetrics.functional.classification import multiclass_hinge_loss
>>> preds = tensor([[0.25, 0.20, 0.55],
... [0.55, 0.05, 0.40],
... [0.10, 0.30, 0.60],
... [0.90, 0.05, 0.05]])
>>> target = tensor([0, 1, 2, 0])
>>> multiclass_hinge_loss(preds, target, num_classes=3)
tensor(0.9125)
>>> multiclass_hinge_loss(preds, target, num_classes=3, squared=True)
tensor(1.1131)
>>> multiclass_hinge_loss(preds, target, num_classes=3, multiclass_mode='one-vs-all')
tensor([0.8750, 1.1250, 1.1000])
"""
if validate_args:
_multiclass_hinge_loss_arg_validation(num_classes, squared, multiclass_mode, ignore_index)
_multiclass_hinge_loss_tensor_validation(preds, target, num_classes, ignore_index)
preds, target = _multiclass_confusion_matrix_format(preds, target, ignore_index, convert_to_labels=False)
measures, total = _multiclass_hinge_loss_update(preds, target, squared, multiclass_mode)
return _hinge_loss_compute(measures, total)
def hinge_loss(
preds: Tensor,
target: Tensor,
task: Literal["binary", "multiclass"],
num_classes: Optional[int] = None,
squared: bool = False,
multiclass_mode: Literal["crammer-singer", "one-vs-all"] = "crammer-singer",
ignore_index: Optional[int] = None,
validate_args: bool = True,
) -> Tensor:
r"""Compute the mean `Hinge loss`_ typically used for Support Vector Machines (SVMs).
This function is a simple wrapper to get the task specific versions of this metric, which is done by setting the
``task`` argument to either ``'binary'`` or ``'multiclass'``. See the documentation of
:func:`~torchmetrics.functional.classification.binary_hinge_loss` and
:func:`~torchmetrics.functional.classification.multiclass_hinge_loss` for the specific details of
each argument influence and examples.
Legacy Example:
>>> from torch import tensor
>>> target = tensor([0, 1, 1])
>>> preds = tensor([0.5, 0.7, 0.1])
>>> hinge_loss(preds, target, task="binary")
tensor(0.9000)
>>> target = tensor([0, 1, 2])
>>> preds = tensor([[-1.0, 0.9, 0.2], [0.5, -1.1, 0.8], [2.2, -0.5, 0.3]])
>>> hinge_loss(preds, target, task="multiclass", num_classes=3)
tensor(1.5551)
>>> target = tensor([0, 1, 2])
>>> preds = tensor([[-1.0, 0.9, 0.2], [0.5, -1.1, 0.8], [2.2, -0.5, 0.3]])
>>> hinge_loss(preds, target, task="multiclass", num_classes=3, multiclass_mode="one-vs-all")
tensor([1.3743, 1.1945, 1.2359])
"""
task = ClassificationTaskNoMultilabel.from_str(task)
if task == ClassificationTaskNoMultilabel.BINARY:
return binary_hinge_loss(preds, target, squared, ignore_index, validate_args)
if task == ClassificationTaskNoMultilabel.MULTICLASS:
if not isinstance(num_classes, int):
raise ValueError(f"`num_classes` is expected to be `int` but `{type(num_classes)} was passed.`")
return multiclass_hinge_loss(preds, target, num_classes, squared, multiclass_mode, ignore_index, validate_args)
raise ValueError(f"Not handled value: {task}")