/
precision_recall.py
552 lines (454 loc) · 25.7 KB
/
precision_recall.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
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
# Copyright The PyTorch 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
from torchmetrics.functional.classification.stat_scores import _reduce_stat_scores, _stat_scores_update
from torchmetrics.utilities.enums import AverageMethod, MDMCAverageMethod
def _precision_compute(
tp: Tensor,
fp: Tensor,
fn: Tensor,
average: str,
mdmc_average: Optional[str],
) -> Tensor:
"""Computes precision from the stat scores: true positives, false positives, true negatives, false negatives.
Args:
tp: True positives
fp: False positives
fn: False negatives
average: Defines the reduction that is applied
mdmc_average: Defines how averaging is done for multi-dimensional multi-class inputs (on top of the
``average`` parameter)
Example:
>>> from torchmetrics.functional.classification.stat_scores import _stat_scores_update
>>> preds = torch.tensor([2, 0, 2, 1])
>>> target = torch.tensor([1, 1, 2, 0])
>>> tp, fp, tn, fn = _stat_scores_update( preds, target, reduce='macro', num_classes=3)
>>> _precision_compute(tp, fp, fn, average='macro', mdmc_average=None)
tensor(0.1667)
>>> tp, fp, tn, fn = _stat_scores_update(preds, target, reduce='micro')
>>> _precision_compute(tp, fp, fn, average='micro', mdmc_average=None)
tensor(0.2500)
"""
numerator = tp.clone()
denominator = tp + fp
if average == AverageMethod.MACRO and mdmc_average != MDMCAverageMethod.SAMPLEWISE:
cond = tp + fp + fn == 0
numerator = numerator[~cond]
denominator = denominator[~cond]
if average == AverageMethod.NONE and mdmc_average != MDMCAverageMethod.SAMPLEWISE:
# a class is not present if there exists no TPs, no FPs, and no FNs
meaningless_indeces = torch.nonzero((tp | fn | fp) == 0).cpu()
numerator[meaningless_indeces, ...] = -1
denominator[meaningless_indeces, ...] = -1
return _reduce_stat_scores(
numerator=numerator,
denominator=denominator,
weights=None if average != "weighted" else tp + fn,
average=average,
mdmc_average=mdmc_average,
)
def precision(
preds: Tensor,
target: Tensor,
average: str = "micro",
mdmc_average: Optional[str] = None,
ignore_index: Optional[int] = None,
num_classes: Optional[int] = None,
threshold: float = 0.5,
top_k: Optional[int] = None,
multiclass: Optional[bool] = None,
) -> Tensor:
r"""Computes `Precision`_
.. math:: \text{Precision} = \frac{\text{TP}}{\text{TP} + \text{FP}}
Where :math:`\text{TP}` and :math:`\text{FP}` represent the number of true positives and
false positives respecitively. With the use of ``top_k`` parameter, this metric can
generalize to Precision@K.
The reduction method (how the precision scores are aggregated) is controlled by the
``average`` parameter, and additionally by the ``mdmc_average`` parameter in the
multi-dimensional multi-class case. Accepts all inputs listed in :ref:`pages/classification:input types`.
Args:
preds: Predictions from model (probabilities, logits or labels)
target: Ground truth values
average:
Defines the reduction that is applied. Should be one of the following:
- ``'micro'`` [default]: Calculate the metric globally, across all samples and classes.
- ``'macro'``: Calculate the metric for each class separately, and average the
metrics across classes (with equal weights for each class).
- ``'weighted'``: Calculate the metric for each class separately, and average the
metrics across classes, weighting each class by its support (``tp + fn``).
- ``'none'`` or ``None``: Calculate the metric for each class separately, and return
the metric for every class.
- ``'samples'``: Calculate the metric for each sample, and average the metrics
across samples (with equal weights for each sample).
.. note:: What is considered a sample in the multi-dimensional multi-class case
depends on the value of ``mdmc_average``.
.. note:: If ``'none'`` and a given class doesn't occur in the ``preds`` or ``target``,
the value for the class will be ``nan``.
mdmc_average:
Defines how averaging is done for multi-dimensional multi-class inputs (on top of the
``average`` parameter). Should be one of the following:
- ``None`` [default]: Should be left unchanged if your data is not multi-dimensional multi-class.
- ``'samplewise'``: In this case, the statistics are computed separately for each
sample on the ``N`` axis, and then averaged over samples.
The computation for each sample is done by treating the flattened extra axes ``...``
(see :ref:`pages/classification:input types`) as the ``N`` dimension within the sample,
and computing the metric for the sample based on that.
- ``'global'``: In this case the ``N`` and ``...`` dimensions of the inputs
(see :ref:`pages/classification:input types`)
are flattened into a new ``N_X`` sample axis, i.e. the inputs are treated as if they
were ``(N_X, C)``. From here on the ``average`` parameter applies as usual.
ignore_index:
Integer specifying a target class to ignore. If given, this class index does not contribute
to the returned score, regardless of reduction method. If an index is ignored, and ``average=None``
or ``'none'``, the score for the ignored class will be returned as ``nan``.
num_classes:
Number of classes. Necessary for ``'macro'``, ``'weighted'`` and ``None`` average methods.
threshold:
Threshold for transforming probability or logit predictions to binary (0,1) predictions, in the case
of binary or multi-label inputs. Default value of 0.5 corresponds to input being probabilities.
top_k:
Number of highest probability or logit score predictions considered to find the correct label,
relevant only for (multi-dimensional) multi-class inputs. The
default value (``None``) will be interpreted as 1 for these inputs.
Should be left at default (``None``) for all other types of inputs.
multiclass:
Used only in certain special cases, where you want to treat inputs as a different type
than what they appear to be. See the parameter's
:ref:`documentation section <pages/classification:using the multiclass parameter>`
for a more detailed explanation and examples.
Return:
The shape of the returned tensor depends on the ``average`` parameter
- If ``average in ['micro', 'macro', 'weighted', 'samples']``, a one-element tensor will be returned
- If ``average in ['none', None]``, the shape will be ``(C,)``, where ``C`` stands for the number of classes
Raises:
ValueError:
If ``average`` is not one of ``"micro"``, ``"macro"``, ``"weighted"``, ``"samples"``, ``"none"`` or ``None``
ValueError:
If ``mdmc_average`` is not one of ``None``, ``"samplewise"``, ``"global"``.
ValueError:
If ``average`` is set but ``num_classes`` is not provided.
ValueError:
If ``num_classes`` is set and ``ignore_index`` is not in the range ``[0, num_classes)``.
Example:
>>> from torchmetrics.functional import precision
>>> preds = torch.tensor([2, 0, 2, 1])
>>> target = torch.tensor([1, 1, 2, 0])
>>> precision(preds, target, average='macro', num_classes=3)
tensor(0.1667)
>>> precision(preds, target, average='micro')
tensor(0.2500)
"""
allowed_average = ["micro", "macro", "weighted", "samples", "none", None]
if average not in allowed_average:
raise ValueError(f"The `average` has to be one of {allowed_average}, got {average}.")
allowed_mdmc_average = [None, "samplewise", "global"]
if mdmc_average not in allowed_mdmc_average:
raise ValueError(f"The `mdmc_average` has to be one of {allowed_mdmc_average}, got {mdmc_average}.")
if average in ["macro", "weighted", "none", None] and (not num_classes or num_classes < 1):
raise ValueError(f"When you set `average` as {average}, you have to provide the number of classes.")
if num_classes and ignore_index is not None and (not 0 <= ignore_index < num_classes or num_classes == 1):
raise ValueError(f"The `ignore_index` {ignore_index} is not valid for inputs with {num_classes} classes")
reduce = "macro" if average in ["weighted", "none", None] else average
tp, fp, _, fn = _stat_scores_update(
preds,
target,
reduce=reduce,
mdmc_reduce=mdmc_average,
threshold=threshold,
num_classes=num_classes,
top_k=top_k,
multiclass=multiclass,
ignore_index=ignore_index,
)
return _precision_compute(tp, fp, fn, average, mdmc_average)
def _recall_compute(
tp: Tensor,
fp: Tensor,
fn: Tensor,
average: str,
mdmc_average: Optional[str],
) -> Tensor:
"""Computes precision from the stat scores: true positives, false positives, true negatives, false negatives.
Args:
tp: True positives
fp: False positives
fn: False negatives
average: Defines the reduction that is applied
mdmc_average: Defines how averaging is done for multi-dimensional multi-class inputs (on top of the
``average`` parameter)
Example:
>>> from torchmetrics.functional.classification.stat_scores import _stat_scores_update
>>> preds = torch.tensor([2, 0, 2, 1])
>>> target = torch.tensor([1, 1, 2, 0])
>>> tp, fp, tn, fn = _stat_scores_update(preds, target, reduce='macro', num_classes=3)
>>> _recall_compute(tp, fp, fn, average='macro', mdmc_average=None)
tensor(0.3333)
>>> tp, fp, tn, fn = _stat_scores_update(preds, target, reduce='micro')
>>> _recall_compute(tp, fp, fn, average='micro', mdmc_average=None)
tensor(0.2500)
"""
numerator = tp.clone()
denominator = tp + fn
if average == AverageMethod.MACRO and mdmc_average != MDMCAverageMethod.SAMPLEWISE:
cond = tp + fp + fn == 0
numerator = numerator[~cond]
denominator = denominator[~cond]
if average == AverageMethod.NONE and mdmc_average != MDMCAverageMethod.SAMPLEWISE:
# a class is not present if there exists no TPs, no FPs, and no FNs
meaningless_indeces = ((tp | fn | fp) == 0).nonzero().cpu()
numerator[meaningless_indeces, ...] = -1
denominator[meaningless_indeces, ...] = -1
return _reduce_stat_scores(
numerator=numerator,
denominator=denominator,
weights=None if average != AverageMethod.WEIGHTED else tp + fn,
average=average,
mdmc_average=mdmc_average,
)
def recall(
preds: Tensor,
target: Tensor,
average: str = "micro",
mdmc_average: Optional[str] = None,
ignore_index: Optional[int] = None,
num_classes: Optional[int] = None,
threshold: float = 0.5,
top_k: Optional[int] = None,
multiclass: Optional[bool] = None,
) -> Tensor:
r"""Computes `Recall`_
.. math:: \text{Recall} = \frac{\text{TP}}{\text{TP} + \text{FN}}
Where :math:`\text{TP}` and :math:`\text{FN}` represent the number of true positives and
false negatives respecitively. With the use of ``top_k`` parameter, this metric can
generalize to Recall@K.
The reduction method (how the recall scores are aggregated) is controlled by the
``average`` parameter, and additionally by the ``mdmc_average`` parameter in the
multi-dimensional multi-class case. Accepts all inputs listed in :ref:`pages/classification:input types`.
Args:
preds: Predictions from model (probabilities, logits or labels)
target: Ground truth values
average:
Defines the reduction that is applied. Should be one of the following:
- ``'micro'`` [default]: Calculate the metric globally, across all samples and classes.
- ``'macro'``: Calculate the metric for each class separately, and average the
metrics across classes (with equal weights for each class).
- ``'weighted'``: Calculate the metric for each class separately, and average the
metrics across classes, weighting each class by its support (``tp + fn``).
- ``'none'`` or ``None``: Calculate the metric for each class separately, and return
the metric for every class.
- ``'samples'``: Calculate the metric for each sample, and average the metrics
across samples (with equal weights for each sample).
.. note:: What is considered a sample in the multi-dimensional multi-class case
depends on the value of ``mdmc_average``.
.. note:: If ``'none'`` and a given class doesn't occur in the ``preds`` or ``target``,
the value for the class will be ``nan``.
mdmc_average:
Defines how averaging is done for multi-dimensional multi-class inputs (on top of the
``average`` parameter). Should be one of the following:
- ``None`` [default]: Should be left unchanged if your data is not multi-dimensional
multi-class.
- ``'samplewise'``: In this case, the statistics are computed separately for each
sample on the ``N`` axis, and then averaged over samples.
The computation for each sample is done by treating the flattened extra axes ``...``
(see :ref:`pages/classification:input types`) as the ``N`` dimension within the sample,
and computing the metric for the sample based on that.
- ``'global'``: In this case the ``N`` and ``...`` dimensions of the inputs
(see :ref:`pages/classification:input types`)
are flattened into a new ``N_X`` sample axis, i.e. the inputs are treated as if they
were ``(N_X, C)``. From here on the ``average`` parameter applies as usual.
ignore_index:
Integer specifying a target class to ignore. If given, this class index does not contribute
to the returned score, regardless of reduction method. If an index is ignored, and ``average=None``
or ``'none'``, the score for the ignored class will be returned as ``nan``.
num_classes:
Number of classes. Necessary for ``'macro'``, ``'weighted'`` and ``None`` average methods.
threshold:
Threshold for transforming probability or logit predictions to binary (0,1) predictions, in the case
of binary or multi-label inputs. Default value of 0.5 corresponds to input being probabilities.
top_k:
Number of the highest probability or logit score predictions considered finding the correct label,
relevant only for (multi-dimensional) multi-class inputs. The
default value (``None``) will be interpreted as 1 for these inputs.
Should be left at default (``None``) for all other types of inputs.
multiclass:
Used only in certain special cases, where you want to treat inputs as a different type
than what they appear to be. See the parameter's
:ref:`documentation section <pages/classification:using the multiclass parameter>`
for a more detailed explanation and examples.
Return:
The shape of the returned tensor depends on the ``average`` parameter
- If ``average in ['micro', 'macro', 'weighted', 'samples']``, a one-element tensor will be returned
- If ``average in ['none', None]``, the shape will be ``(C,)``, where ``C`` stands for the number of classes
Raises:
ValueError:
If ``average`` is not one of ``"micro"``, ``"macro"``, ``"weighted"``, ``"samples"``, ``"none"`` or ``None``
ValueError:
If ``mdmc_average`` is not one of ``None``, ``"samplewise"``, ``"global"``.
ValueError:
If ``average`` is set but ``num_classes`` is not provided.
ValueError:
If ``num_classes`` is set and ``ignore_index`` is not in the range ``[0, num_classes)``.
Example:
>>> from torchmetrics.functional import recall
>>> preds = torch.tensor([2, 0, 2, 1])
>>> target = torch.tensor([1, 1, 2, 0])
>>> recall(preds, target, average='macro', num_classes=3)
tensor(0.3333)
>>> recall(preds, target, average='micro')
tensor(0.2500)
"""
allowed_average = ("micro", "macro", "weighted", "samples", "none", None)
if average not in allowed_average:
raise ValueError(f"The `average` has to be one of {allowed_average}, got {average}.")
allowed_mdmc_average = (None, "samplewise", "global")
if mdmc_average not in allowed_mdmc_average:
raise ValueError(f"The `mdmc_average` has to be one of {allowed_mdmc_average}, got {mdmc_average}.")
if average in ["macro", "weighted", "none", None] and (not num_classes or num_classes < 1):
raise ValueError(f"When you set `average` as {average}, you have to provide the number of classes.")
if num_classes and ignore_index is not None and (not 0 <= ignore_index < num_classes or num_classes == 1):
raise ValueError(f"The `ignore_index` {ignore_index} is not valid for inputs with {num_classes} classes")
reduce = "macro" if average in ["weighted", "none", None] else average
tp, fp, _, fn = _stat_scores_update(
preds,
target,
reduce=reduce,
mdmc_reduce=mdmc_average,
threshold=threshold,
num_classes=num_classes,
top_k=top_k,
multiclass=multiclass,
ignore_index=ignore_index,
)
return _recall_compute(tp, fp, fn, average, mdmc_average)
def precision_recall(
preds: Tensor,
target: Tensor,
average: str = "micro",
mdmc_average: Optional[str] = None,
ignore_index: Optional[int] = None,
num_classes: Optional[int] = None,
threshold: float = 0.5,
top_k: Optional[int] = None,
multiclass: Optional[bool] = None,
) -> Tuple[Tensor, Tensor]:
r"""Computes `Precision`_
.. math:: \text{Precision} = \frac{\text{TP}}{\text{TP} + \text{FP}}
.. math:: \text{Recall} = \frac{\text{TP}}{\text{TP} + \text{FN}}
Where :math:`\text{TP}`m :math:`\text{FN}` and :math:`\text{FP}` represent the number
of true positives, false negatives and false positives respecitively. With the use of
``top_k`` parameter, this metric can generalize to Recall@K and Precision@K.
The reduction method (how the recall scores are aggregated) is controlled by the
``average`` parameter, and additionally by the ``mdmc_average`` parameter in the
multi-dimensional multi-class case. Accepts all inputs listed in :ref:`pages/classification:input types`.
Args:
preds: Predictions from model (probabilities, logits or labels)
target: Ground truth values
average:
Defines the reduction that is applied. Should be one of the following:
- ``'micro'`` [default]: Calculate the metric globally, across all samples and classes.
- ``'macro'``: Calculate the metric for each class separately, and average the
metrics across classes (with equal weights for each class).
- ``'weighted'``: Calculate the metric for each class separately, and average the
metrics across classes, weighting each class by its support (``tp + fn``).
- ``'none'`` or ``None``: Calculate the metric for each class separately, and return
the metric for every class.
- ``'samples'``: Calculate the metric for each sample, and average the metrics
across samples (with equal weights for each sample).
.. note:: What is considered a sample in the multi-dimensional multi-class case
depends on the value of ``mdmc_average``.
.. note:: If ``'none'`` and a given class doesn't occur in the ``preds`` or ``target``,
the value for the class will be ``nan``.
mdmc_average:
Defines how averaging is done for multi-dimensional multi-class inputs (on top of the
``average`` parameter). Should be one of the following:
- ``None`` [default]: Should be left unchanged if your data is not multi-dimensional multi-class.
- ``'samplewise'``: In this case, the statistics are computed separately for each
sample on the ``N`` axis, and then averaged over samples.
The computation for each sample is done by treating the flattened extra axes ``...``
(see :ref:`pages/classification:input types`) as the ``N`` dimension within the sample,
and computing the metric for the sample based on that.
- ``'global'``: In this case the ``N`` and ``...`` dimensions of the inputs
(see :ref:`pages/classification:input types`)
are flattened into a new ``N_X`` sample axis, i.e. the inputs are treated as if they
were ``(N_X, C)``. From here on the ``average`` parameter applies as usual.
ignore_index:
Integer specifying a target class to ignore. If given, this class index does not contribute
to the returned score, regardless of reduction method. If an index is ignored, and ``average=None``
or ``'none'``, the score for the ignored class will be returned as ``nan``.
num_classes:
Number of classes. Necessary for ``'macro'``, ``'weighted'`` and ``None`` average methods.
threshold:
Threshold for transforming probability or logit predictions to binary (0,1) predictions, in the case
of binary or multi-label inputs. Default value of 0.5 corresponds to input being probabilities.
top_k:
Number of highest probability or logit score predictions considered to find the correct label,
relevant only for (multi-dimensional) multi-class inputs. The
default value (``None``) will be interpreted as 1 for these inputs.
Should be left at default (``None``) for all other types of inputs.
multiclass:
Used only in certain special cases, where you want to treat inputs as a different type
than what they appear to be. See the parameter's
:ref:`documentation section <pages/classification:using the multiclass parameter>`
for a more detailed explanation and examples.
Return:
The function returns a tuple with two elements: precision and recall. Their shape
depends on the ``average`` parameter
- If ``average in ['micro', 'macro', 'weighted', 'samples']``, they are a single element tensor
- If ``average in ['none', None]``, they are a tensor of shape ``(C, )``, where ``C`` stands for
the number of classes
Raises:
ValueError:
If ``average`` is not one of ``"micro"``, ``"macro"``, ``"weighted"``, ``"samples"``, ``"none"`` or ``None``
ValueError:
If ``mdmc_average`` is not one of ``None``, ``"samplewise"``, ``"global"``.
ValueError:
If ``average`` is set but ``num_classes`` is not provided.
ValueError:
If ``num_classes`` is set and ``ignore_index`` is not in the range ``[0, num_classes)``.
Example:
>>> from torchmetrics.functional import precision_recall
>>> preds = torch.tensor([2, 0, 2, 1])
>>> target = torch.tensor([1, 1, 2, 0])
>>> precision_recall(preds, target, average='macro', num_classes=3)
(tensor(0.1667), tensor(0.3333))
>>> precision_recall(preds, target, average='micro')
(tensor(0.2500), tensor(0.2500))
"""
allowed_average = ("micro", "macro", "weighted", "samples", "none", None)
if average not in allowed_average:
raise ValueError(f"The `average` has to be one of {allowed_average}, got {average}.")
allowed_mdmc_average = (None, "samplewise", "global")
if mdmc_average not in allowed_mdmc_average:
raise ValueError(f"The `mdmc_average` has to be one of {allowed_mdmc_average}, got {mdmc_average}.")
if average in ["macro", "weighted", "none", None] and (not num_classes or num_classes < 1):
raise ValueError(f"When you set `average` as {average}, you have to provide the number of classes.")
if num_classes and ignore_index is not None and (not 0 <= ignore_index < num_classes or num_classes == 1):
raise ValueError(f"The `ignore_index` {ignore_index} is not valid for inputs with {num_classes} classes")
reduce = "macro" if average in ["weighted", "none", None] else average
tp, fp, _, fn = _stat_scores_update(
preds,
target,
reduce=reduce,
mdmc_reduce=mdmc_average,
threshold=threshold,
num_classes=num_classes,
top_k=top_k,
multiclass=multiclass,
ignore_index=ignore_index,
)
precision_ = _precision_compute(tp, fp, fn, average, mdmc_average)
recall_ = _recall_compute(tp, fp, fn, average, mdmc_average)
return precision_, recall_