guide.rst 22.8 KB
Newer Older
André Anjos's avatar
André Anjos committed
1 2 3 4
.. vim: set fileencoding=utf-8 :
.. Andre Anjos <andre.dos.anjos@gmail.com>
.. Tue 15 Oct 17:41:52 2013

André Anjos's avatar
André Anjos committed
5
.. testsetup:: *
André Anjos's avatar
André Anjos committed
6

André Anjos's avatar
André Anjos committed
7 8 9 10 11 12
  import numpy
  positives = numpy.random.normal(1,1,100)
  negatives = numpy.random.normal(-1,1,100)
  import matplotlib
  if not hasattr(matplotlib, 'backends'):
    matplotlib.use('pdf') #non-interactive avoids exception on display
André Anjos's avatar
André Anjos committed
13
  import bob.measure
André Anjos's avatar
André Anjos committed
14 15 16 17 18

============
 User Guide
============

André Anjos's avatar
André Anjos committed
19
Methods in the :py:mod:`bob.measure` module can help you to quickly and easily
20 21
evaluate error for multi-class or binary classification problems. If you are
not yet familiarized with aspects of performance evaluation, we recommend the
22 23
following papers and book chapters for an overview of some of the implemented
methods.
24 25 26 27 28 29 30

* Bengio, S., Keller, M., Mariéthoz, J. (2004). `The Expected Performance
  Curve`_.  International Conference on Machine Learning ICML Workshop on ROC
  Analysis in Machine Learning, 136(1), 1963–1966.
* Martin, A., Doddington, G., Kamm, T., Ordowski, M., & Przybocki, M. (1997).
  `The DET curve in assessment of detection task performance`_. Fifth European
  Conference on Speech Communication and Technology (pp. 1895-1898).
31 32
* Li, S., Jain, A.K. (2005), `Handbook of Face Recognition`, Chapter 14, Springer

33 34 35 36 37 38 39 40 41 42 43 44 45

Overview
--------

A classifier is subject to two types of errors, either the real access/signal
is rejected (false rejection) or an impostor attack/a false access is accepted
(false acceptance). A possible way to measure the detection performance is to
use the Half Total Error Rate (HTER), which combines the False Rejection Rate
(FRR) and the False Acceptance Rate (FAR) and is defined in the following
formula:

.. math::

46
   HTER(\tau, \mathcal{D}) = \frac{FAR(\tau, \mathcal{D}) + FRR(\tau, \mathcal{D})}{2} \quad \textrm{[\%]}
47

André Anjos's avatar
André Anjos committed
48

49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66
where :math:`\mathcal{D}` denotes the dataset used. Since both the FAR and the
FRR depends on the threshold :math:`\tau`, they are strongly related to each
other: increasing the FAR will reduce the FRR and vice-versa. For this reason,
results are often presented using either a Receiver Operating Characteristic
(ROC) or a Detection-Error Tradeoff (DET) plot, these two plots basically
present the FAR versus the FRR for different values of the threshold. Another
widely used measure to summarise the performance of a system is the Equal Error
Rate (EER), defined as the point along the ROC or DET curve where the FAR
equals the FRR.

However, it was noted in by Bengio et al. (2004) that ROC and DET curves may be
misleading when comparing systems. Hence, the so-called Expected Performance
Curve (EPC) was proposed and consists of an unbiased estimate of the reachable
performance of a system at various operating points.  Indeed, in real-world
scenarios, the threshold :math:`\tau` has to be set a priori: this is typically
done using a development set (also called cross-validation set). Nevertheless,
the optimal threshold can be different depending on the relative importance
given to the FAR and the FRR. Hence, in the EPC framework, the cost
67 68
:math:`\beta \in [0;1]` is defined as the trade-off between the FAR and FRR.
The optimal threshold :math:`\tau^*` is then computed using different values of
69 70 71 72 73
:math:`\beta`, corresponding to different operating points:

.. math::
  \tau^{*} = \arg\!\min_{\tau} \quad \beta \cdot \textrm{FAR}(\tau, \mathcal{D}_{d}) + (1-\beta) \cdot \textrm{FRR}(\tau, \mathcal{D}_{d})

André Anjos's avatar
André Anjos committed
74

75
where :math:`\mathcal{D}_{d}` denotes the development set and should be
André Anjos's avatar
André Anjos committed
76
completely separate to the evaluation set :math:`\mathcal{D}`.
77 78 79 80 81

Performance for different values of :math:`\beta` is then computed on the test
set :math:`\mathcal{D}_{t}` using the previously derived threshold. Note that
setting :math:`\beta` to 0.5 yields to the Half Total Error Rate (HTER) as
defined in the first equation.
André Anjos's avatar
André Anjos committed
82 83 84

.. note::

85
  Most of the methods available in this module require as input a set of 2
86 87 88 89 90 91
  :py:class:`numpy.ndarray` objects that contain the scores obtained by the
  classification system to be evaluated, without specific order. Most of the
  classes that are defined to deal with two-class problems. Therefore, in this
  setting, and throughout this manual, we have defined that the **negatives**
  represents the impostor attacks or false class accesses (that is when a
  sample of class A is given to the classifier of another class, such as class
92
  B) for of the classifier. The second set, referred as the **positives**
93 94 95 96 97 98 99 100 101
  represents the true class accesses or signal response of the classifier. The
  vectors are called this way because the procedures implemented in this module
  expects that the scores of **negatives** to be statistically distributed to
  the left of the signal scores (the **positives**). If that is not the case,
  one should either invert the input to the methods or multiply all scores
  available by -1, in order to have them inverted.

  The input to create these two vectors is generated by experiments conducted
  by the user and normally sits in files that may need some parsing before
102 103 104 105 106 107
  these vectors can be extracted. While it is not possible to provide a parser 
  for every individual file that may be generated in different experimental
  frameworks, we do provide a parser for a generic two columns format
  where the first column contains -1/1 for negative/positive and the second column
  contains score values. Please refer to the documentation of 
  :py:func:`bob.measure.load.split` for more details.
108

109
  In the remainder of this section we assume you have successfully parsed and
110 111
  loaded your scores in two 1D float64 vectors and are ready to evaluate the
  performance of the classifier.
André Anjos's avatar
André Anjos committed
112

113 114
Verification
------------
André Anjos's avatar
André Anjos committed
115

116 117
To count the number of correctly classified positives and negatives you can use
the following techniques:
André Anjos's avatar
André Anjos committed
118 119 120

.. doctest::

121 122 123 124 125 126
   >>> # negatives, positives = parse_my_scores(...) # write parser if not provided!
   >>> T = 0.0 #Threshold: later we explain how one can calculate these
   >>> correct_negatives = bob.measure.correctly_classified_negatives(negatives, T)
   >>> FAR = 1 - (float(correct_negatives.sum())/negatives.size)
   >>> correct_positives = bob.measure.correctly_classified_positives(positives, T)
   >>> FRR = 1 - (float(correct_positives.sum())/positives.size)
André Anjos's avatar
André Anjos committed
127

128
We do provide a method to calculate the FAR and FRR in a single shot:
André Anjos's avatar
André Anjos committed
129 130 131

.. doctest::

132
   >>> FAR, FRR = bob.measure.farfrr(negatives, positives, T)
André Anjos's avatar
André Anjos committed
133

134 135 136 137 138 139
The threshold ``T`` is normally calculated by looking at the distribution of
negatives and positives in a development (or validation) set, selecting a
threshold that matches a certain criterion and applying this derived threshold
to the test (or evaluation) set. This technique gives a better overview of the
generalization of a method. We implement different techniques for the
calculation of the threshold:
André Anjos's avatar
André Anjos committed
140

141
* Threshold for the EER
André Anjos's avatar
André Anjos committed
142

143
  .. doctest::
André Anjos's avatar
André Anjos committed
144

André Anjos's avatar
André Anjos committed
145
    >>> T = bob.measure.eer_threshold(negatives, positives)
André Anjos's avatar
André Anjos committed
146

147
* Threshold for the minimum HTER
André Anjos's avatar
André Anjos committed
148

149
  .. doctest::
André Anjos's avatar
André Anjos committed
150

André Anjos's avatar
André Anjos committed
151
    >>> T = bob.measure.min_hter_threshold(negatives, positives)
André Anjos's avatar
André Anjos committed
152

153 154
* Threshold for the minimum weighted error rate (MWER) given a certain cost
  :math:`\beta`.
André Anjos's avatar
André Anjos committed
155

156
  .. doctest:: python
André Anjos's avatar
André Anjos committed
157

158
     >>> cost = 0.3 #or "beta"
André Anjos's avatar
André Anjos committed
159
     >>> T = bob.measure.min_weighted_error_rate_threshold(negatives, positives, cost)
André Anjos's avatar
André Anjos committed
160

161
  .. note::
André Anjos's avatar
André Anjos committed
162

163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178
     By setting cost to 0.5 is equivalent to use
     :py:func:`bob.measure.min_hter_threshold`.

.. note::
   Many functions in ``bob.measure`` have an ``is_sorted`` parameter, which defaults to ``False``, throughout.
   However, these functions need sorted ``positive`` and/or ``negative`` scores.
   If scores are not in ascendantly sorted order, internally, they will be copied -- twice!
   To avoid scores to be copied, you might want to sort the scores in ascending order, e.g., by:

   .. doctest:: python

      >>> negatives.sort()
      >>> positives.sort()
      >>> t = bob.measure.min_weighted_error_rate_threshold(negatives, positives, cost, is_sorted = True)
      >>> assert T == t

179 180 181 182 183 184 185 186 187 188 189 190 191 192 193
Identification
--------------

For identification, the Recognition Rate is one of the standard measures.
To compute recognition rates, you can use the :py:func:`bob.measure.recognition_rate` function.
This function expects a relatively complex data structure, which is the same as for the `CMC`_ below.
For each probe item, the scores for negative and positive comparisons are computed, and collected for all probe items:

.. doctest::

   >>> rr_scores = []
   >>> for probe in range(10):
   ...   pos = numpy.random.normal(1, 1, 1)
   ...   neg = numpy.random.normal(0, 1, 19)
   ...   rr_scores.append((neg, pos))
194
   >>> rr = bob.measure.recognition_rate(rr_scores, rank=1)
195 196 197 198 199 200 201 202 203 204 205 206 207 208 209

For open set identification, according to Li and Jain (2005) there are two different error measures defined.
The first measure is the :py:func:`bob.measure.detection_identification_rate`, which counts the number of correctly classified in-gallery probe items.
The second measure is the :py:func:`bob.measure.false_alarm_rate`, which counts, how often an out-of-gallery probe item was incorrectly accepted.
Both rates can be computed using the same data structure, with one exception.
Both functions require that at least one probe item exists, which has no according gallery item, i.e., where the positives are empty or ``None``:

(continued from above...)

.. doctest::

   >>> for probe in range(10):
   ...   pos = None
   ...   neg = numpy.random.normal(-2, 1, 10)
   ...   rr_scores.append((neg, pos))
210 211
   >>> dir = bob.measure.detection_identification_rate(rr_scores, threshold = 0, rank=1)
   >>> far = bob.measure.false_alarm_rate(rr_scores, threshold = 0)
212

213 214 215
Confidence interval
-------------------

216 217 218 219 220 221 222 223 224 225
A confidence interval for parameter :math:`x` consists of a lower
estimate :math:`L`, and an upper estimate :math:`U`, such that the probability
of the true value being within the interval estimate is equal to :math:`\alpha`.
For example, a 95% confidence interval (i.e. :math:`\alpha = 0.95`) for a
parameter :math:`x` is given by :math:`[L, U]` such that

.. math:: Prob(x∈[L,U]) = 95% 

The smaller the test size, the wider the confidence 
interval will be, and the greater :math:`\alpha`, the smaller the confidence interval
226 227 228 229 230
will be.

`The Clopper-Pearson interval`_, a common method for calculating
confidence intervals, is function of the number of success, the number of trials 
and confidence
231
value :math:`\alpha` is used as :py:func:`bob.measure.utils.confidence_for_indicator_variable`.
232 233
It is based on the cumulative probabilities of the binomial distribution. This
method is quite conservative, meaning that the true coverage rate of a 95% 
234
Clopper–Pearson interval may be well above 95%.
235

236 237 238 239 240 241 242 243 244 245 246
For example, we want to evaluate the reliability of a system to
identify registered persons. Let's say that among 10,000 accepted
transactions, 9856 are true matches. The 95% confidence interval for true match
rate is then:
.. doctest:: python

    >>> bob.measure.utils.confidence_for_indicator_variable(9856, 10000)
    (0.98306835053282549, 0.98784270928084694)

meaning there is a 95% probability that the true match rate is inside :math:`[0.983,
0.988]`.
André Anjos's avatar
André Anjos committed
247

248 249
Plotting
--------
André Anjos's avatar
André Anjos committed
250

251
An image is worth 1000 words, they say. You can combine the capabilities of
252 253
`Matplotlib`_ with |project| to plot a number of curves. However, you must have
that package installed though. In this section we describe a few recipes.
André Anjos's avatar
André Anjos committed
254

255 256
ROC
===
André Anjos's avatar
André Anjos committed
257

258 259 260
The Receiver Operating Characteristic (ROC) curve is one of the oldest plots in
town. To plot an ROC curve, in possession of your **negatives** and
**positives**, just do something along the lines of:
André Anjos's avatar
André Anjos committed
261 262 263

.. doctest::

264 265 266 267 268 269 270 271
   >>> from matplotlib import pyplot
   >>> # we assume you have your negatives and positives already split
   >>> npoints = 100
   >>> bob.measure.plot.roc(negatives, positives, npoints, color=(0,0,0), linestyle='-', label='test') # doctest: +SKIP
   >>> pyplot.xlabel('FAR (%)') # doctest: +SKIP
   >>> pyplot.ylabel('FRR (%)') # doctest: +SKIP
   >>> pyplot.grid(True)
   >>> pyplot.show() # doctest: +SKIP
André Anjos's avatar
André Anjos committed
272

273
You should see an image like the following one:
André Anjos's avatar
André Anjos committed
274

André Anjos's avatar
André Anjos committed
275 276 277
.. plot::

   import numpy
278
   numpy.random.seed(42)
André Anjos's avatar
André Anjos committed
279
   import bob.measure
André Anjos's avatar
André Anjos committed
280 281 282 283 284
   from matplotlib import pyplot

   positives = numpy.random.normal(1,1,100)
   negatives = numpy.random.normal(-1,1,100)
   npoints = 100
André Anjos's avatar
André Anjos committed
285
   bob.measure.plot.roc(negatives, positives, npoints, color=(0,0,0), linestyle='-', label='test')
André Anjos's avatar
André Anjos committed
286 287 288 289
   pyplot.grid(True)
   pyplot.xlabel('FAR (%)')
   pyplot.ylabel('FRR (%)')
   pyplot.title('ROC')
André Anjos's avatar
André Anjos committed
290

291
As can be observed, plotting methods live in the namespace
292 293 294 295 296 297 298 299 300 301 302 303
:py:mod:`bob.measure.plot`. They work like the
:py:func:`matplotlib.pyplot.plot` itself, except that instead of receiving the
x and y point coordinates as parameters, they receive the two
:py:class:`numpy.ndarray` arrays with negatives and positives, as well as an
indication of the number of points the curve must contain.

As in the :py:func:`matplotlib.pyplot.plot` command, you can pass optional
parameters for the line as shown in the example to setup its color, shape and
even the label.  For an overview of the keywords accepted, please refer to the
`Matplotlib`_'s Documentation. Other plot properties such as the plot title,
axis labels, grids, legends should be controlled directly using the relevant
`Matplotlib`_'s controls.
André Anjos's avatar
André Anjos committed
304

305 306
DET
===
André Anjos's avatar
André Anjos committed
307

308
A DET curve can be drawn using similar commands such as the ones for the ROC curve:
André Anjos's avatar
André Anjos committed
309 310 311

.. doctest::

312 313 314
  >>> from matplotlib import pyplot
  >>> # we assume you have your negatives and positives already split
  >>> npoints = 100
André Anjos's avatar
André Anjos committed
315 316
  >>> bob.measure.plot.det(negatives, positives, npoints, color=(0,0,0), linestyle='-', label='test') # doctest: +SKIP
  >>> bob.measure.plot.det_axis([0.01, 40, 0.01, 40]) # doctest: +SKIP
317 318 319 320
  >>> pyplot.xlabel('FAR (%)') # doctest: +SKIP
  >>> pyplot.ylabel('FRR (%)') # doctest: +SKIP
  >>> pyplot.grid(True)
  >>> pyplot.show() # doctest: +SKIP
André Anjos's avatar
André Anjos committed
321

322
This will produce an image like the following one:
André Anjos's avatar
André Anjos committed
323

André Anjos's avatar
André Anjos committed
324 325 326
.. plot::

   import numpy
327
   numpy.random.seed(42)
André Anjos's avatar
André Anjos committed
328
   import bob.measure
André Anjos's avatar
André Anjos committed
329 330 331 332 333 334
   from matplotlib import pyplot

   positives = numpy.random.normal(1,1,100)
   negatives = numpy.random.normal(-1,1,100)

   npoints = 100
André Anjos's avatar
André Anjos committed
335 336
   bob.measure.plot.det(negatives, positives, npoints, color=(0,0,0), linestyle='-', label='test')
   bob.measure.plot.det_axis([0.1, 80, 0.1, 80])
André Anjos's avatar
André Anjos committed
337 338 339 340
   pyplot.grid(True)
   pyplot.xlabel('FAR (%)')
   pyplot.ylabel('FRR (%)')
   pyplot.title('DET')
André Anjos's avatar
André Anjos committed
341 342 343

.. note::

344 345 346
  If you wish to reset axis zooming, you must use the Gaussian scale rather
  than the visual marks showed at the plot, which are just there for
  displaying purposes. The real axis scale is based on the
347
  :py:func:`bob.measure.ppndf` method. For example, if you wish to set the x and y
348
  axis to display data between 1% and 40% here is the recipe:
André Anjos's avatar
André Anjos committed
349

350
  .. doctest::
André Anjos's avatar
André Anjos committed
351

352
    >>> #AFTER you plot the DET curve, just set the axis in this way:
André Anjos's avatar
André Anjos committed
353
    >>> pyplot.axis([bob.measure.ppndf(k/100.0) for k in (1, 40, 1, 40)]) # doctest: +SKIP
André Anjos's avatar
André Anjos committed
354

355
  We provide a convenient way for you to do the above in this module. So,
André Anjos's avatar
André Anjos committed
356
  optionally, you may use the ``bob.measure.plot.det_axis`` method like this:
André Anjos's avatar
André Anjos committed
357

358
  .. doctest::
André Anjos's avatar
André Anjos committed
359

André Anjos's avatar
André Anjos committed
360
    >>> bob.measure.plot.det_axis([1, 40, 1, 40]) # doctest: +SKIP
André Anjos's avatar
André Anjos committed
361

362 363
EPC
===
André Anjos's avatar
André Anjos committed
364

365
Drawing an EPC requires that both the development set negatives and positives are provided alongside
366
the test (or evaluation) set ones. Because of this the API is slightly modified:
André Anjos's avatar
André Anjos committed
367 368 369

.. doctest::

André Anjos's avatar
André Anjos committed
370
  >>> bob.measure.plot.epc(dev_neg, dev_pos, test_neg, test_pos, npoints, color=(0,0,0), linestyle='-') # doctest: +SKIP
371
  >>> pyplot.show() # doctest: +SKIP
André Anjos's avatar
André Anjos committed
372

373
This will produce an image like the following one:
André Anjos's avatar
André Anjos committed
374

André Anjos's avatar
André Anjos committed
375 376 377
.. plot::

   import numpy
378
   numpy.random.seed(42)
André Anjos's avatar
André Anjos committed
379
   import bob.measure
André Anjos's avatar
André Anjos committed
380 381 382 383 384 385 386
   from matplotlib import pyplot

   dev_pos = numpy.random.normal(1,1,100)
   dev_neg = numpy.random.normal(-1,1,100)
   test_pos = numpy.random.normal(0.9,1,100)
   test_neg = numpy.random.normal(-1.1,1,100)
   npoints = 100
André Anjos's avatar
André Anjos committed
387
   bob.measure.plot.epc(dev_neg, dev_pos, test_neg, test_pos, npoints, color=(0,0,0), linestyle='-')
André Anjos's avatar
André Anjos committed
388 389
   pyplot.grid(True)
   pyplot.title('EPC')
André Anjos's avatar
André Anjos committed
390

391 392 393 394

CMC
===

395 396 397 398 399
The Cumulative Match Characteristics (CMC) curve estimates the probability that
the correct model is in the *N* models with the highest similarity to a given
probe.  A CMC curve can be plotted using the :py:func:`bob.measure.plot.cmc`
function.  The CMC can be calculated from a relatively complex data structure,
which defines a pair of positive and negative scores **per probe**:
400 401 402 403

.. plot::

   import numpy
404
   numpy.random.seed(42)
405 406 407
   import bob.measure
   from matplotlib import pyplot

408
   cmc_scores = []
409 410 411
   for probe in range(10):
     positives = numpy.random.normal(1, 1, 1)
     negatives = numpy.random.normal(0, 1, 19)
412 413
     cmc_scores.append((negatives, positives))
   bob.measure.plot.cmc(cmc_scores, logx=False)
André Anjos's avatar
André Anjos committed
414
   pyplot.grid(True)
415 416 417 418 419
   pyplot.title('CMC')
   pyplot.xlabel('Rank')
   pyplot.xticks([1,5,10,20])
   pyplot.xlim([1,20])
   pyplot.ylim([0,100])
420
   pyplot.ylabel('Probability of Recognition (%)')
421 422 423

Usually, there is only a single positive score per probe, but this is not a fixed restriction.

424 425 426 427

Detection & Identification Curve
================================

André Anjos's avatar
André Anjos committed
428 429 430 431 432 433 434 435
The detection & identification curve is designed to evaluate open set
identification tasks.  It can be plotted using the
:py:func:`bob.measure.plot.detection_identification_curve` function, but it
requires at least one open-set probe, i.e., where no corresponding positive
score exists, for which the FAR values are computed.  Here, we plot the
detection and identification curve for rank 1, so that the recognition rate for
FAR=1 will be identical to the rank one :py:func:`bob.measure.recognition_rate`
obtained in the CMC plot above.
436 437 438 439 440 441 442 443 444

.. plot::

   import numpy
   numpy.random.seed(42)
   import bob.measure
   from matplotlib import pyplot

   cmc_scores = []
445
   for probe in range(1000):
446 447 448
     positives = numpy.random.normal(1, 1, 1)
     negatives = numpy.random.normal(0, 1, 19)
     cmc_scores.append((negatives, positives))
449
   for probe in range(1000):
450 451 452 453 454
     negatives = numpy.random.normal(-1, 1, 10)
     cmc_scores.append((negatives, None))

   bob.measure.plot.detection_identification_curve(cmc_scores, rank=1, logx=True)
   pyplot.xlabel('False Alarm Rate')
455
   pyplot.xlim([0.0001, 1])
456
   pyplot.ylabel('Detection & Identification Rate (%)')
457
   pyplot.ylim([0,1])
458 459 460



461 462
Fine-tunning
============
André Anjos's avatar
André Anjos committed
463

André Anjos's avatar
André Anjos committed
464 465
The methods inside :py:mod:`bob.measure.plot` are only provided as a
`Matplotlib`_ wrapper to equivalent methods in :py:mod:`bob.measure` that can
466 467
only calculate the points without doing any plotting. You may prefer to tweak
the plotting or even use a different plotting system such as gnuplot. Have a
André Anjos's avatar
André Anjos committed
468 469 470 471
look at the implementations at :py:mod:`bob.measure.plot` to understand how to
use the |project| methods to compute the curves and interlace that in the way
that best suits you.

472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489
Full applications
-----------------

Commands under ``bob measure`` can be used to quickly evaluate a set of
scores and generate plots. We present these commands in this section. The commands
take as input generic 2-column data format as specified in the documentation of
:py:func:`bob.measure.load.split`

Metrics
=======

To calculate the threshold using a certain criterion (EER (default) or min.HTER)
on a set, after setting up |project|, just do:

.. code-block:: sh

  $ bob measure metrics dev-1.txt
  [Min. criterion: EER] Threshold on Development set `dev-1.txt`: -8.025286e-03
490 491 492 493 494 495 496 497 498
  ====  ===================
  ..    Development dev-1
  ====  ===================
  FMR   6.263% (31/495)
  FNMR  6.208% (28/451)
  FAR   5.924%
  FRR   11.273%
  HTER  8.599%
  ====  ===================
499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516

The output will present the threshold together with the FMR, FMNR, FAR, FRR and
HTER on the given set, calculated using such a threshold. The relative counts of FAs
and FRs are also displayed between parenthesis.

.. note::
    Several scores files can be given at once and the metrics will be computed
    for each of them separatly. Development and test files must be given by
    pairs and the ``--test`` (or ``-t``) flag must be given (otherwise test 
    scores are treated as development scores)


To evaluate the performance of a new score file with a given threshold, use
``--thres``:

.. code-block:: sh

  $ bob measure metrics --thres 0.006 test-1.txt
517 518 519 520 521 522 523 524 525 526
  [Min. criterion: user provider] Threshold on Development set `test-1`: 6.000000e-03
  ====  ====================
  ..    Development test-1
  ====  ====================
  FMR   5.010% (24/479)
  FNMR  6.977% (33/473)
  FAR   4.770%
  FRR   11.442%
  HTER  8.106%
  ====  ====================
527 528 529 530 531 532 533

You can simultaneously conduct the threshold computation and its performance
on a test set:

.. code-block:: sh

  $ bob measure metrics --test dev-1.txt test-1.txt
534 535 536 537 538 539 540 541 542 543
  [Min. criterion: EER] Threshold on Development set `dev-1`: -8.025286e-03
  ====  ===================  ===============
  ..    Development dev-1    Test test-1
  ====  ===================  ===============
  FMR   6.263% (31/495)      5.637% (27/479)
  FNMR  6.208% (28/451)      6.131% (29/473)
  FAR   5.924%               5.366%
  FRR   11.273%              10.637%
  HTER  8.599%               8.001%
  ====  ===================  ===============
544 545 546 547

.. note::
    Table format can be changed using ``--tablefmt`` option. Note that
    ``latex``
548 549
    format is used by default when the outputs are witten in a log file and
    ``rst`` when written in the terminal. See
550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600
    ``bob measure metrics --help`` for more details.
    

Plots
=====

Customizable plotting commands are available in the :py:mod:`bob.measure` module.
They take a list of development and/or test files and generate a single PDF
file containing the plots. Available plots are:

*  ``roc`` (receiver operating characteristic)

*  ``det`` (detection error trade-off)

*  ``epc`` (expected performance curve)

*  ``hist`` (histograms of positive and negatives)

Use the ``--help`` option on the above-cited commands to find-out about more
options.

For example, to generate a DET curve from development and test datasets:

.. code-block:: sh

    $bob measure det --test --split --output 'my_det.pdf' dev-1.txt test-1.txt 
    dev-2.txt test-2.txt

where `my_det.pdf` will contain DET plots for the two experiments.

.. note::
    By default, ``det`` and ``roc`` plot development and evaluation curves on
    different plots. You can force gather everything in the same plot using
    ``--no-split`` option.

Evaluate
========

A convenient command `evaluate` is provided to generate multiple metrics and
plots for a list of experiments. It generates two `metrics` outputs with EER
and HTER criteria along with `roc`, `det`, `epc`, `hist` plots for each
experiment. For example:

.. code-block:: sh

    $bob measure evaluate -t -l 'my_metrics.txt' -o 'my_plots.pdf' {sys1, sys2}/
    {test,dev}

will output metrics and plots for the two experiments (dev and test pairs) in
`my_metrics.txt` and `my_plots.pdf`, respectively.

André Anjos's avatar
André Anjos committed
601 602
.. include:: links.rst

603
.. Place youre references here:
André Anjos's avatar
André Anjos committed
604

605 606
.. _`The Expected Performance Curve`: http://publications.idiap.ch/downloads/reports/2005/bengio_2005_icml.pdf
.. _`The DET curve in assessment of detection task performance`: http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.117.4489&rep=rep1&type=pdf
607
.. _`The Clopper-Pearson interval`: https://en.wikipedia.org/wiki/Binomial_proportion_confidence_interval#Clopper-Pearson_interval