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
|
<!DOCTYPE html>
<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
<head>
<meta charset="utf-8">
<meta http-equiv="X-UA-Compatible" content="IE=edge">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="canonical" href="http://keras.io/preprocessing/sequence/">
<link rel="shortcut icon" href="../../img/favicon.ico">
<title>Sequence Preprocessing - Keras Documentation</title>
<link href='https://fonts.googleapis.com/css?family=Lato:400,700|Source+Sans+Pro:400,700|Inconsolata:400,700' rel='stylesheet' type='text/css'>
<link rel="stylesheet" href="../../css/theme.css" type="text/css" />
<link rel="stylesheet" href="../../css/theme_extra.css" type="text/css" />
<link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/highlight.js/9.12.0/styles/github.min.css">
<script>
// Current page data
var mkdocs_page_name = "Sequence Preprocessing";
var mkdocs_page_input_path = "preprocessing/sequence.md";
var mkdocs_page_url = "/preprocessing/sequence/";
</script>
<script src="../../js/jquery-2.1.1.min.js" defer></script>
<script src="../../js/modernizr-2.8.3.min.js" defer></script>
<script src="https://cdnjs.cloudflare.com/ajax/libs/highlight.js/9.12.0/highlight.min.js"></script>
<script>hljs.initHighlightingOnLoad();</script>
<script>
(function(i,s,o,g,r,a,m){i['GoogleAnalyticsObject']=r;i[r]=i[r]||function(){
(i[r].q=i[r].q||[]).push(arguments)},i[r].l=1*new Date();a=s.createElement(o),
m=s.getElementsByTagName(o)[0];a.async=1;a.src=g;m.parentNode.insertBefore(a,m)
})(window,document,'script','https://www.google-analytics.com/analytics.js','ga');
ga('create', 'UA-61785484-1', 'keras.io');
ga('send', 'pageview');
</script>
</head>
<body class="wy-body-for-nav" role="document">
<div class="wy-grid-for-nav">
<nav data-toggle="wy-nav-shift" class="wy-nav-side stickynav">
<div class="wy-side-scroll">
<a href="">
<div class="keras-logo">
<img src="/img/keras-logo-small.jpg" class="keras-logo-img">
Keras Documentation
</div>
</a>
<div class="wy-side-nav-search">
<div role="search">
<form id ="rtd-search-form" class="wy-form" action="../../search.html" method="get">
<input type="text" name="q" placeholder="Search docs" title="Type search term here" />
</form>
</div>
</div>
<div class="wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
<ul>
<li class="toctree-l1"><a class="reference internal" href="../..">Home</a>
</li>
</ul>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../why-use-keras/">Why use Keras</a>
</li>
</ul>
<p class="caption"><span class="caption-text">Getting started</span></p>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../getting-started/sequential-model-guide/">Guide to the Sequential model</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../getting-started/functional-api-guide/">Guide to the Functional API</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../getting-started/faq/">FAQ</a>
</li>
</ul>
<p class="caption"><span class="caption-text">Models</span></p>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../models/about-keras-models/">About Keras models</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../models/sequential/">Sequential</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../models/model/">Model (functional API)</a>
</li>
</ul>
<p class="caption"><span class="caption-text">Layers</span></p>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../layers/about-keras-layers/">About Keras layers</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../layers/core/">Core Layers</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../layers/convolutional/">Convolutional Layers</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../layers/pooling/">Pooling Layers</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../layers/local/">Locally-connected Layers</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../layers/recurrent/">Recurrent Layers</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../layers/embeddings/">Embedding Layers</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../layers/merge/">Merge Layers</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../layers/advanced-activations/">Advanced Activations Layers</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../layers/normalization/">Normalization Layers</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../layers/noise/">Noise layers</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../layers/wrappers/">Layer wrappers</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../layers/writing-your-own-keras-layers/">Writing your own Keras layers</a>
</li>
</ul>
<p class="caption"><span class="caption-text">Preprocessing</span></p>
<ul class="current">
<li class="toctree-l1 current"><a class="reference internal current" href="./">Sequence Preprocessing</a>
<ul class="current">
</ul>
</li>
<li class="toctree-l1"><a class="reference internal" href="../text/">Text Preprocessing</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../image/">Image Preprocessing</a>
</li>
</ul>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../losses/">Losses</a>
</li>
</ul>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../metrics/">Metrics</a>
</li>
</ul>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../optimizers/">Optimizers</a>
</li>
</ul>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../activations/">Activations</a>
</li>
</ul>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../callbacks/">Callbacks</a>
</li>
</ul>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../datasets/">Datasets</a>
</li>
</ul>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../applications/">Applications</a>
</li>
</ul>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../backend/">Backend</a>
</li>
</ul>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../initializers/">Initializers</a>
</li>
</ul>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../regularizers/">Regularizers</a>
</li>
</ul>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../constraints/">Constraints</a>
</li>
</ul>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../visualization/">Visualization</a>
</li>
</ul>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../scikit-learn-api/">Scikit-learn API</a>
</li>
</ul>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../utils/">Utils</a>
</li>
</ul>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../contributing/">Contributing</a>
</li>
</ul>
<p class="caption"><span class="caption-text">Examples</span></p>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../examples/addition_rnn/">Addition RNN</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../examples/antirectifier/">Custom layer - antirectifier</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../examples/babi_rnn/">Baby RNN</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../examples/babi_memnn/">Baby MemNN</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../examples/cifar10_cnn/">CIFAR-10 CNN</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../examples/cifar10_resnet/">CIFAR-10 ResNet</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../examples/conv_filter_visualization/">Convolution filter visualization</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../examples/conv_lstm/">Convolutional LSTM</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../examples/deep_dream/">Deep Dream</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../examples/image_ocr/">Image OCR</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../examples/imdb_bidirectional_lstm/">Bidirectional LSTM</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../examples/imdb_cnn/">1D CNN for text classification</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../examples/imdb_cnn_lstm/">Sentiment classification CNN-LSTM</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../examples/imdb_fasttext/">Fasttext for text classification</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../examples/imdb_lstm/">Sentiment classification LSTM</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../examples/lstm_seq2seq/">Sequence to sequence - training</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../examples/lstm_seq2seq_restore/">Sequence to sequence - prediction</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../examples/lstm_stateful/">Stateful LSTM</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../examples/lstm_text_generation/">LSTM for text generation</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../examples/mnist_acgan/">Auxiliary Classifier GAN</a>
</li>
</ul>
</div>
</div>
</nav>
<section data-toggle="wy-nav-shift" class="wy-nav-content-wrap">
<nav class="wy-nav-top" role="navigation" aria-label="top navigation">
<i data-toggle="wy-nav-top" class="fa fa-bars"></i>
<a href="../..">Keras Documentation</a>
</nav>
<div class="wy-nav-content">
<div class="rst-content">
<div role="navigation" aria-label="breadcrumbs navigation">
<ul class="wy-breadcrumbs">
<li><a href="../..">Docs</a> »</li>
<li>Preprocessing »</li>
<li>Sequence Preprocessing</li>
<li class="wy-breadcrumbs-aside">
<a href="https://github.com/keras-team/keras/tree/master/docs"
class="icon icon-github"> Edit on GitHub</a>
</li>
</ul>
<hr/>
</div>
<div role="main">
<div class="section">
<p><span style="float:right;"><a href="https://github.com/keras-team/keras/blob/master/keras/preprocessing/sequence.py#L16">[source]</a></span></p>
<h3 id="timeseriesgenerator">TimeseriesGenerator</h3>
<pre><code class="python">keras.preprocessing.sequence.TimeseriesGenerator(data, targets, length, sampling_rate=1, stride=1, start_index=0, end_index=None, shuffle=False, reverse=False, batch_size=128)
</code></pre>
<p>Utility class for generating batches of temporal data.</p>
<p>This class takes in a sequence of data-points gathered at
equal intervals, along with time series parameters such as
stride, length of history, etc., to produce batches for
training/validation.</p>
<p><strong>Arguments</strong></p>
<ul>
<li><strong>data</strong>: Indexable generator (such as list or Numpy array)
containing consecutive data points (timesteps).
The data should be at 2D, and axis 0 is expected
to be the time dimension.</li>
<li><strong>targets</strong>: Targets corresponding to timesteps in <code>data</code>.
It should have same length as <code>data</code>.</li>
<li><strong>length</strong>: Length of the output sequences (in number of timesteps).</li>
<li><strong>sampling_rate</strong>: Period between successive individual timesteps
within sequences. For rate <code>r</code>, timesteps
<code>data[i]</code>, <code>data[i-r]</code>, ... <code>data[i - length]</code>
are used for create a sample sequence.</li>
<li><strong>stride</strong>: Period between successive output sequences.
For stride <code>s</code>, consecutive output samples would
be centered around <code>data[i]</code>, <code>data[i+s]</code>, <code>data[i+2*s]</code>, etc.</li>
<li><strong>start_index</strong>: Data points earlier than <code>start_index</code> will not be used
in the output sequences. This is useful to reserve part of the
data for test or validation.</li>
<li><strong>end_index</strong>: Data points later than <code>end_index</code> will not be used
in the output sequences. This is useful to reserve part of the
data for test or validation.</li>
<li><strong>shuffle</strong>: Whether to shuffle output samples,
or instead draw them in chronological order.</li>
<li><strong>reverse</strong>: Boolean: if <code>true</code>, timesteps in each output sample will be
in reverse chronological order.</li>
<li><strong>batch_size</strong>: Number of timeseries samples in each batch
(except maybe the last one).</li>
</ul>
<p><strong>Returns</strong></p>
<p>A <a href="/utils/#sequence">Sequence</a> instance.</p>
<p><strong>Examples</strong></p>
<pre><code class="python">from keras.preprocessing.sequence import TimeseriesGenerator
import numpy as np
data = np.array([[i] for i in range(50)])
targets = np.array([[i] for i in range(50)])
data_gen = TimeseriesGenerator(data, targets,
length=10, sampling_rate=2,
batch_size=2)
assert len(data_gen) == 20
batch_0 = data_gen[0]
x, y = batch_0
assert np.array_equal(x,
np.array([[[0], [2], [4], [6], [8]],
[[1], [3], [5], [7], [9]]]))
assert np.array_equal(y,
np.array([[10], [11]]))
</code></pre>
<hr />
<h3 id="pad_sequences">pad_sequences</h3>
<pre><code class="python">keras.preprocessing.sequence.pad_sequences(sequences, maxlen=None, dtype='int32', padding='pre', truncating='pre', value=0.0)
</code></pre>
<p>Pads sequences to the same length.</p>
<p>This function transforms a list of
<code>num_samples</code> sequences (lists of integers)
into a 2D Numpy array of shape <code>(num_samples, num_timesteps)</code>.
<code>num_timesteps</code> is either the <code>maxlen</code> argument if provided,
or the length of the longest sequence otherwise.</p>
<p>Sequences that are shorter than <code>num_timesteps</code>
are padded with <code>value</code> at the end.</p>
<p>Sequences longer than <code>num_timesteps</code> are truncated
so that they fit the desired length.
The position where padding or truncation happens is determined by
the arguments <code>padding</code> and <code>truncating</code>, respectively.</p>
<p>Pre-padding is the default.</p>
<p><strong>Arguments</strong></p>
<ul>
<li><strong>sequences</strong>: List of lists, where each element is a sequence.</li>
<li><strong>maxlen</strong>: Int, maximum length of all sequences.</li>
<li><strong>dtype</strong>: Type of the output sequences.
To pad sequences with variable length strings, you can use <code>object</code>.</li>
<li><strong>padding</strong>: String, 'pre' or 'post':
pad either before or after each sequence.</li>
<li><strong>truncating</strong>: String, 'pre' or 'post':
remove values from sequences larger than
<code>maxlen</code>, either at the beginning or at the end of the sequences.</li>
<li><strong>value</strong>: Float or String, padding value.</li>
</ul>
<p><strong>Returns</strong></p>
<ul>
<li><strong>x</strong>: Numpy array with shape <code>(len(sequences), maxlen)</code></li>
</ul>
<p><strong>Raises</strong></p>
<ul>
<li><strong>ValueError</strong>: In case of invalid values for <code>truncating</code> or <code>padding</code>,
or in case of invalid shape for a <code>sequences</code> entry.</li>
</ul>
<hr />
<h3 id="skipgrams">skipgrams</h3>
<pre><code class="python">keras.preprocessing.sequence.skipgrams(sequence, vocabulary_size, window_size=4, negative_samples=1.0, shuffle=True, categorical=False, sampling_table=None, seed=None)
</code></pre>
<p>Generates skipgram word pairs.</p>
<p>This function transforms a sequence of word indexes (list of integers)
into tuples of words of the form:</p>
<ul>
<li>(word, word in the same window), with label 1 (positive samples).</li>
<li>(word, random word from the vocabulary), with label 0 (negative samples).</li>
</ul>
<p>Read more about Skipgram in this gnomic paper by Mikolov et al.:
<a href="http://arxiv.org/pdf/1301.3781v3.pdf">Efficient Estimation of Word Representations in
Vector Space</a></p>
<p><strong>Arguments</strong></p>
<ul>
<li><strong>sequence</strong>: A word sequence (sentence), encoded as a list
of word indices (integers). If using a <code>sampling_table</code>,
word indices are expected to match the rank
of the words in a reference dataset (e.g. 10 would encode
the 10-th most frequently occurring token).
Note that index 0 is expected to be a non-word and will be skipped.</li>
<li><strong>vocabulary_size</strong>: Int, maximum possible word index + 1</li>
<li><strong>window_size</strong>: Int, size of sampling windows (technically half-window).
The window of a word <code>w_i</code> will be
<code>[i - window_size, i + window_size+1]</code>.</li>
<li><strong>negative_samples</strong>: Float >= 0. 0 for no negative (i.e. random) samples.
1 for same number as positive samples.</li>
<li><strong>shuffle</strong>: Whether to shuffle the word couples before returning them.</li>
<li><strong>categorical</strong>: bool. if False, labels will be
integers (eg. <code>[0, 1, 1 .. ]</code>),
if <code>True</code>, labels will be categorical, e.g.
<code>[[1,0],[0,1],[0,1] .. ]</code>.</li>
<li><strong>sampling_table</strong>: 1D array of size <code>vocabulary_size</code> where the entry i
encodes the probability to sample a word of rank i.</li>
<li><strong>seed</strong>: Random seed.</li>
</ul>
<p><strong>Returns</strong></p>
<p>couples, labels: where <code>couples</code> are int pairs and
<code>labels</code> are either 0 or 1.</p>
<p><strong>Note</strong></p>
<p>By convention, index 0 in the vocabulary is
a non-word and will be skipped.</p>
<hr />
<h3 id="make_sampling_table">make_sampling_table</h3>
<pre><code class="python">keras.preprocessing.sequence.make_sampling_table(size, sampling_factor=1e-05)
</code></pre>
<p>Generates a word rank-based probabilistic sampling table.</p>
<p>Used for generating the <code>sampling_table</code> argument for <code>skipgrams</code>.
<code>sampling_table[i]</code> is the probability of sampling
the word i-th most common word in a dataset
(more common words should be sampled less frequently, for balance).</p>
<p>The sampling probabilities are generated according
to the sampling distribution used in word2vec:</p>
<pre><code>p(word) = (min(1, sqrt(word_frequency / sampling_factor) /
(word_frequency / sampling_factor)))
</code></pre>
<p>We assume that the word frequencies follow Zipf's law (s=1) to derive
a numerical approximation of frequency(rank):</p>
<p><code>frequency(rank) ~ 1/(rank * (log(rank) + gamma) + 1/2 - 1/(12*rank))</code>
where <code>gamma</code> is the Euler-Mascheroni constant.</p>
<p><strong>Arguments</strong></p>
<ul>
<li><strong>size</strong>: Int, number of possible words to sample.</li>
<li><strong>sampling_factor</strong>: The sampling factor in the word2vec formula.</li>
</ul>
<p><strong>Returns</strong></p>
<p>A 1D Numpy array of length <code>size</code> where the ith entry
is the probability that a word of rank i should be sampled.</p>
</div>
</div>
<footer>
<div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
<a href="../text/" class="btn btn-neutral float-right" title="Text Preprocessing">Next <span class="icon icon-circle-arrow-right"></span></a>
<a href="../../layers/writing-your-own-keras-layers/" class="btn btn-neutral" title="Writing your own Keras layers"><span class="icon icon-circle-arrow-left"></span> Previous</a>
</div>
<hr/>
<div role="contentinfo">
<!-- Copyright etc -->
</div>
Built with <a href="https://www.mkdocs.org/">MkDocs</a> using a <a href="https://github.com/snide/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
</footer>
</div>
</div>
</section>
</div>
<div class="rst-versions" role="note" aria-label="versions">
<span class="rst-current-version" data-toggle="rst-current-version">
<a href="http://github.com/keras-team/keras/" class="fa fa-github" style="float: left; color: #fcfcfc"> GitHub</a>
<span><a href="../../layers/writing-your-own-keras-layers/" style="color: #fcfcfc;">« Previous</a></span>
<span style="margin-left: 15px"><a href="../text/" style="color: #fcfcfc">Next »</a></span>
</span>
</div>
<script>var base_url = '../..';</script>
<script src="../../js/theme.js" defer></script>
<script src="../../search/main.js" defer></script>
<script type="text/javascript" defer>
window.onload = function () {
SphinxRtdTheme.Navigation.enable(true);
};
</script>
</body>
</html>
|
