-
Notifications
You must be signed in to change notification settings - Fork 81
/
Copy pathplot_johnson_lindenstrauss_bound.html
473 lines (414 loc) · 45.2 KB
/
plot_johnson_lindenstrauss_bound.html
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
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
<title>The Johnson-Lindenstrauss bound for embedding with random projections — scikit-learn 0.19.2 documentation</title>
<!-- htmltitle is before nature.css - we use this hack to load bootstrap first -->
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
<link rel="stylesheet" href="../_static/css/bootstrap.min.css" media="screen" />
<link rel="stylesheet" href="../_static/css/bootstrap-responsive.css"/>
<link rel="stylesheet" href="../_static/nature.css" type="text/css" />
<link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
<link rel="stylesheet" href="../_static/gallery.css" type="text/css" />
<script type="text/javascript">
var DOCUMENTATION_OPTIONS = {
URL_ROOT: '../',
VERSION: '0.19.2',
COLLAPSE_INDEX: false,
FILE_SUFFIX: '.html',
HAS_SOURCE: true,
SOURCELINK_SUFFIX: '.txt'
};
</script>
<script type="text/javascript" src="../_static/jquery.js"></script>
<script type="text/javascript" src="../_static/underscore.js"></script>
<script type="text/javascript" src="../_static/doctools.js"></script>
<script type="text/javascript" src="../_static/js/copybutton.js"></script>
<link rel="shortcut icon" href="../_static/favicon.ico"/>
<link rel="author" title="About these documents" href="../about.html" />
<link rel="search" title="Search" href="../search.html" />
<link rel="next" title="Comparison of kernel ridge regression and SVR" href="plot_kernel_ridge_regression.html" />
<link rel="prev" title="Multilabel classification" href="plot_multilabel.html" />
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
<script src="../_static/js/bootstrap.min.js" type="text/javascript"></script>
<script>
VERSION_SUBDIR = (function(groups) {
return groups ? groups[1] : null;
})(location.href.match(/^https?:\/\/scikit-learn.org\/([^\/]+)/));
</script>
<link rel="canonical" href="https://scikit-learn.org/stable/auto_examples/plot_johnson_lindenstrauss_bound.html" />
<script type="text/javascript">
$("div.buttonNext, div.buttonPrevious").hover(
function () {
$(this).css('background-color', '#FF9C34');
},
function () {
$(this).css('background-color', '#A7D6E2');
}
);
function showMenu() {
var topNav = document.getElementById("scikit-navbar");
if (topNav.className === "navbar") {
topNav.className += " responsive";
} else {
topNav.className = "navbar";
}
};
</script>
</head>
<body>
<div class="header-wrapper">
<div class="header">
<p class="logo"><a href="../index.html">
<img src="../_static/scikit-learn-logo-small.png" alt="Logo"/>
</a>
</p><div class="navbar" id="scikit-navbar">
<ul>
<li><a href="../index.html">Home</a></li>
<li><a href="../install.html">Installation</a></li>
<li class="btn-li"><div class="btn-group">
<a href="../documentation.html">Documentation</a>
<a class="btn dropdown-toggle" data-toggle="dropdown">
<span class="caret"></span>
</a>
<ul class="dropdown-menu">
<li class="link-title">Scikit-learn <script>document.write(DOCUMENTATION_OPTIONS.VERSION + (VERSION_SUBDIR ? " (" + VERSION_SUBDIR + ")" : ""));</script></li>
<li><a href="../tutorial/index.html">Tutorials</a></li>
<li><a href="../user_guide.html">User guide</a></li>
<li><a href="../modules/classes.html">API</a></li>
<li><a href="../faq.html">FAQ</a></li>
<li><a href="../developers/contributing.html">Contributing</a></li>
<li class="divider"></li>
<script>if (VERSION_SUBDIR != "stable") document.write('<li><a href="http://scikit-learn.org/stable/documentation.html">Stable version</a></li>')</script>
<script>if (VERSION_SUBDIR != "dev") document.write('<li><a href="http://scikit-learn.org/dev/documentation.html">Development version</a></li>')</script>
<li><a href="http://scikit-learn.org/dev/versions.html">All available versions</a></li>
<li><a href="../_downloads/scikit-learn-docs.pdf">PDF documentation</a></li>
</ul>
</div>
</li>
<li><a href="index.html">Examples</a></li>
</ul>
<a href="javascript:void(0);" onclick="showMenu()">
<div class="nav-icon">
<div class="hamburger-line"></div>
<div class="hamburger-line"></div>
<div class="hamburger-line"></div>
</div>
</a>
<div class="search_form">
<div class="gcse-search" id="cse" style="width: 100%;"></div>
</div>
</div> <!-- end navbar --></div>
</div>
<!-- GitHub "fork me" ribbon -->
<a href="https://github.com/scikit-learn/scikit-learn">
<img class="fork-me"
style="position: absolute; top: 0; right: 0; border: 0;"
src="../_static/img/forkme.png"
alt="Fork me on GitHub" />
</a>
<div class="content-wrapper">
<div class="sphinxsidebar">
<div class="sphinxsidebarwrapper">
<div class="rel">
<div class="rellink">
<a href="plot_multilabel.html"
accesskey="P">Previous
<br/>
<span class="smallrellink">
Multilabel cl...
</span>
<span class="hiddenrellink">
Multilabel classification
</span>
</a>
</div>
<div class="spacer">
</div>
<div class="rellink">
<a href="plot_kernel_ridge_regression.html"
accesskey="N">Next
<br/>
<span class="smallrellink">
Comparison of...
</span>
<span class="hiddenrellink">
Comparison of kernel ridge regression and SVR
</span>
</a>
</div>
<!-- Ad a link to the 'up' page -->
<div class="spacer">
</div>
<div class="rellink">
<a href="index.html">
Up
<br/>
<span class="smallrellink">
Examples
</span>
<span class="hiddenrellink">
Examples
</span>
</a>
</div>
</div>
<p class="doc-version"><b>scikit-learn v0.19.2</b><br/>
<a href="http://scikit-learn.org/stable/support.html#documentation-resources">Other versions</a></p>
<p class="citing">Please <b><a href="../about.html#citing-scikit-learn" style="font-size: 110%;">cite us </a></b>if you use the software.</p>
<ul>
<li><a class="reference internal" href="#">The Johnson-Lindenstrauss bound for embedding with random projections</a><ul>
<li><a class="reference internal" href="#theoretical-bounds">Theoretical bounds</a></li>
<li><a class="reference internal" href="#empirical-validation">Empirical validation</a></li>
<li><a class="reference internal" href="#remarks">Remarks</a></li>
</ul>
</li>
</ul>
</div>
</div>
<input type="checkbox" id="nav-trigger" class="nav-trigger" checked />
<label for="nav-trigger"></label>
<div class="content">
<div class="documentwrapper">
<div class="bodywrapper">
<div class="body" role="main">
<div class="section" id="the-johnson-lindenstrauss-bound-for-embedding-with-random-projections">
<span id="sphx-glr-auto-examples-plot-johnson-lindenstrauss-bound-py"></span><h1>The Johnson-Lindenstrauss bound for embedding with random projections<a class="headerlink" href="#the-johnson-lindenstrauss-bound-for-embedding-with-random-projections" title="Permalink to this headline">¶</a></h1>
<p>The <a class="reference external" href="https://en.wikipedia.org/wiki/Johnson%E2%80%93Lindenstrauss_lemma">Johnson-Lindenstrauss lemma</a> states that any high dimensional
dataset can be randomly projected into a lower dimensional Euclidean
space while controlling the distortion in the pairwise distances.</p>
<div class="section" id="theoretical-bounds">
<h2>Theoretical bounds<a class="headerlink" href="#theoretical-bounds" title="Permalink to this headline">¶</a></h2>
<p>The distortion introduced by a random projection <cite>p</cite> is asserted by
the fact that <cite>p</cite> is defining an eps-embedding with good probability
as defined by:</p>
<div class="math">
<p><img src="../_images/math/f959343b770e80df34786eebb08e80e60d69d91f.png" alt="(1 - eps) \|u - v\|^2 < \|p(u) - p(v)\|^2 < (1 + eps) \|u - v\|^2"/></p>
</div><p>Where u and v are any rows taken from a dataset of shape [n_samples,
n_features] and p is a projection by a random Gaussian N(0, 1) matrix
with shape [n_components, n_features] (or a sparse Achlioptas matrix).</p>
<p>The minimum number of components to guarantees the eps-embedding is
given by:</p>
<div class="math">
<p><img src="../_images/math/065d63cc8b891d9269d412a42f0691aeaec0f49f.png" alt="n\_components >= 4 log(n\_samples) / (eps^2 / 2 - eps^3 / 3)"/></p>
</div><p>The first plot shows that with an increasing number of samples <code class="docutils literal"><span class="pre">n_samples</span></code>,
the minimal number of dimensions <code class="docutils literal"><span class="pre">n_components</span></code> increased logarithmically
in order to guarantee an <code class="docutils literal"><span class="pre">eps</span></code>-embedding.</p>
<p>The second plot shows that an increase of the admissible
distortion <code class="docutils literal"><span class="pre">eps</span></code> allows to reduce drastically the minimal number of
dimensions <code class="docutils literal"><span class="pre">n_components</span></code> for a given number of samples <code class="docutils literal"><span class="pre">n_samples</span></code></p>
</div>
<div class="section" id="empirical-validation">
<h2>Empirical validation<a class="headerlink" href="#empirical-validation" title="Permalink to this headline">¶</a></h2>
<p>We validate the above bounds on the digits dataset or on the 20 newsgroups
text document (TF-IDF word frequencies) dataset:</p>
<ul class="simple">
<li>for the digits dataset, some 8x8 gray level pixels data for 500
handwritten digits pictures are randomly projected to spaces for various
larger number of dimensions <code class="docutils literal"><span class="pre">n_components</span></code>.</li>
<li>for the 20 newsgroups dataset some 500 documents with 100k
features in total are projected using a sparse random matrix to smaller
euclidean spaces with various values for the target number of dimensions
<code class="docutils literal"><span class="pre">n_components</span></code>.</li>
</ul>
<p>The default dataset is the digits dataset. To run the example on the twenty
newsgroups dataset, pass the –twenty-newsgroups command line argument to this
script.</p>
<p>For each value of <code class="docutils literal"><span class="pre">n_components</span></code>, we plot:</p>
<ul class="simple">
<li>2D distribution of sample pairs with pairwise distances in original
and projected spaces as x and y axis respectively.</li>
<li>1D histogram of the ratio of those distances (projected / original).</li>
</ul>
<p>We can see that for low values of <code class="docutils literal"><span class="pre">n_components</span></code> the distribution is wide
with many distorted pairs and a skewed distribution (due to the hard
limit of zero ratio on the left as distances are always positives)
while for larger values of n_components the distortion is controlled
and the distances are well preserved by the random projection.</p>
</div>
<div class="section" id="remarks">
<h2>Remarks<a class="headerlink" href="#remarks" title="Permalink to this headline">¶</a></h2>
<p>According to the JL lemma, projecting 500 samples without too much distortion
will require at least several thousands dimensions, irrespective of the
number of features of the original dataset.</p>
<p>Hence using random projections on the digits dataset which only has 64 features
in the input space does not make sense: it does not allow for dimensionality
reduction in this case.</p>
<p>On the twenty newsgroups on the other hand the dimensionality can be decreased
from 56436 down to 10000 while reasonably preserving pairwise distances.</p>
<ul class="sphx-glr-horizontal">
<li><a class="first reference internal image-reference" href="../_images/sphx_glr_plot_johnson_lindenstrauss_bound_001.png"><img alt="../_images/sphx_glr_plot_johnson_lindenstrauss_bound_001.png" src="../_images/sphx_glr_plot_johnson_lindenstrauss_bound_001.png" style="width: 300.79999999999995px; height: 225.6px;" /></a>
</li>
<li><a class="first reference internal image-reference" href="../_images/sphx_glr_plot_johnson_lindenstrauss_bound_002.png"><img alt="../_images/sphx_glr_plot_johnson_lindenstrauss_bound_002.png" src="../_images/sphx_glr_plot_johnson_lindenstrauss_bound_002.png" style="width: 300.79999999999995px; height: 225.6px;" /></a>
</li>
<li><a class="first reference internal image-reference" href="../_images/sphx_glr_plot_johnson_lindenstrauss_bound_003.png"><img alt="../_images/sphx_glr_plot_johnson_lindenstrauss_bound_003.png" src="../_images/sphx_glr_plot_johnson_lindenstrauss_bound_003.png" style="width: 300.79999999999995px; height: 225.6px;" /></a>
</li>
<li><a class="first reference internal image-reference" href="../_images/sphx_glr_plot_johnson_lindenstrauss_bound_004.png"><img alt="../_images/sphx_glr_plot_johnson_lindenstrauss_bound_004.png" src="../_images/sphx_glr_plot_johnson_lindenstrauss_bound_004.png" style="width: 300.79999999999995px; height: 225.6px;" /></a>
</li>
<li><a class="first reference internal image-reference" href="../_images/sphx_glr_plot_johnson_lindenstrauss_bound_005.png"><img alt="../_images/sphx_glr_plot_johnson_lindenstrauss_bound_005.png" src="../_images/sphx_glr_plot_johnson_lindenstrauss_bound_005.png" style="width: 300.79999999999995px; height: 225.6px;" /></a>
</li>
<li><a class="first reference internal image-reference" href="../_images/sphx_glr_plot_johnson_lindenstrauss_bound_006.png"><img alt="../_images/sphx_glr_plot_johnson_lindenstrauss_bound_006.png" src="../_images/sphx_glr_plot_johnson_lindenstrauss_bound_006.png" style="width: 300.79999999999995px; height: 225.6px;" /></a>
</li>
<li><a class="first reference internal image-reference" href="../_images/sphx_glr_plot_johnson_lindenstrauss_bound_007.png"><img alt="../_images/sphx_glr_plot_johnson_lindenstrauss_bound_007.png" src="../_images/sphx_glr_plot_johnson_lindenstrauss_bound_007.png" style="width: 300.79999999999995px; height: 225.6px;" /></a>
</li>
<li><a class="first reference internal image-reference" href="../_images/sphx_glr_plot_johnson_lindenstrauss_bound_008.png"><img alt="../_images/sphx_glr_plot_johnson_lindenstrauss_bound_008.png" src="../_images/sphx_glr_plot_johnson_lindenstrauss_bound_008.png" style="width: 300.79999999999995px; height: 225.6px;" /></a>
</li>
</ul>
<p class="sphx-glr-script-out">Out:</p>
<div class="sphx-glr-script-out highlight-default"><div class="highlight"><pre><span></span><span class="n">Embedding</span> <span class="mi">500</span> <span class="n">samples</span> <span class="k">with</span> <span class="n">dim</span> <span class="mi">64</span> <span class="n">using</span> <span class="n">various</span> <span class="n">random</span> <span class="n">projections</span>
<span class="n">Projected</span> <span class="mi">500</span> <span class="n">samples</span> <span class="kn">from</span> <span class="mi">64</span> <span class="n">to</span> <span class="mi">300</span> <span class="ow">in</span> <span class="mf">0.018</span><span class="n">s</span>
<span class="n">Random</span> <span class="n">matrix</span> <span class="k">with</span> <span class="n">size</span><span class="p">:</span> <span class="mf">0.029</span><span class="n">MB</span>
<span class="n">Mean</span> <span class="n">distances</span> <span class="n">rate</span><span class="p">:</span> <span class="mf">1.01</span> <span class="p">(</span><span class="mf">0.09</span><span class="p">)</span>
<span class="n">Projected</span> <span class="mi">500</span> <span class="n">samples</span> <span class="kn">from</span> <span class="mi">64</span> <span class="n">to</span> <span class="mi">1000</span> <span class="ow">in</span> <span class="mf">0.066</span><span class="n">s</span>
<span class="n">Random</span> <span class="n">matrix</span> <span class="k">with</span> <span class="n">size</span><span class="p">:</span> <span class="mf">0.098</span><span class="n">MB</span>
<span class="n">Mean</span> <span class="n">distances</span> <span class="n">rate</span><span class="p">:</span> <span class="mf">1.03</span> <span class="p">(</span><span class="mf">0.05</span><span class="p">)</span>
<span class="n">Projected</span> <span class="mi">500</span> <span class="n">samples</span> <span class="kn">from</span> <span class="mi">64</span> <span class="n">to</span> <span class="mi">10000</span> <span class="ow">in</span> <span class="mf">0.430</span><span class="n">s</span>
<span class="n">Random</span> <span class="n">matrix</span> <span class="k">with</span> <span class="n">size</span><span class="p">:</span> <span class="mf">0.960</span><span class="n">MB</span>
<span class="n">Mean</span> <span class="n">distances</span> <span class="n">rate</span><span class="p">:</span> <span class="mf">1.00</span> <span class="p">(</span><span class="mf">0.01</span><span class="p">)</span>
</pre></div>
</div>
<div class="line-block">
<div class="line"><br /></div>
</div>
<div class="highlight-python"><div class="highlight"><pre><span></span><span class="k">print</span><span class="p">(</span><span class="vm">__doc__</span><span class="p">)</span>
<span class="kn">import</span> <span class="nn">sys</span>
<span class="kn">from</span> <span class="nn">time</span> <span class="kn">import</span> <span class="n">time</span>
<span class="kn">import</span> <span class="nn">numpy</span> <span class="kn">as</span> <span class="nn">np</span>
<span class="kn">import</span> <span class="nn">matplotlib.pyplot</span> <span class="kn">as</span> <span class="nn">plt</span>
<span class="kn">from</span> <span class="nn">sklearn.random_projection</span> <span class="kn">import</span> <a href="../modules/generated/sklearn.random_projection.johnson_lindenstrauss_min_dim.html#sklearn.random_projection.johnson_lindenstrauss_min_dim" title="View documentation for sklearn.random_projection.johnson_lindenstrauss_min_dim"><span class="n">johnson_lindenstrauss_min_dim</span></a>
<span class="kn">from</span> <span class="nn">sklearn.random_projection</span> <span class="kn">import</span> <a href="../modules/generated/sklearn.random_projection.SparseRandomProjection.html#sklearn.random_projection.SparseRandomProjection" title="View documentation for sklearn.random_projection.SparseRandomProjection"><span class="n">SparseRandomProjection</span></a>
<span class="kn">from</span> <span class="nn">sklearn.datasets</span> <span class="kn">import</span> <a href="../modules/generated/sklearn.datasets.fetch_20newsgroups_vectorized.html#sklearn.datasets.fetch_20newsgroups_vectorized" title="View documentation for sklearn.datasets.fetch_20newsgroups_vectorized"><span class="n">fetch_20newsgroups_vectorized</span></a>
<span class="kn">from</span> <span class="nn">sklearn.datasets</span> <span class="kn">import</span> <a href="../modules/generated/sklearn.datasets.load_digits.html#sklearn.datasets.load_digits" title="View documentation for sklearn.datasets.load_digits"><span class="n">load_digits</span></a>
<span class="kn">from</span> <span class="nn">sklearn.metrics.pairwise</span> <span class="kn">import</span> <span class="n">euclidean_distances</span>
<span class="c1"># Part 1: plot the theoretical dependency between n_components_min and</span>
<span class="c1"># n_samples</span>
<span class="c1"># range of admissible distortions</span>
<span class="n">eps_range</span> <span class="o">=</span> <a href="http://docs.scipy.org/doc/numpy-1.8.1/reference/generated/numpy.linspace.html#numpy.linspace" title="View documentation for numpy.linspace"><span class="n">np</span><span class="o">.</span><span class="n">linspace</span></a><span class="p">(</span><span class="mf">0.1</span><span class="p">,</span> <span class="mf">0.99</span><span class="p">,</span> <span class="mi">5</span><span class="p">)</span>
<span class="n">colors</span> <span class="o">=</span> <span class="n">plt</span><span class="o">.</span><span class="n">cm</span><span class="o">.</span><span class="n">Blues</span><span class="p">(</span><a href="http://docs.scipy.org/doc/numpy-1.8.1/reference/generated/numpy.linspace.html#numpy.linspace" title="View documentation for numpy.linspace"><span class="n">np</span><span class="o">.</span><span class="n">linspace</span></a><span class="p">(</span><span class="mf">0.3</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">eps_range</span><span class="p">)))</span>
<span class="c1"># range of number of samples (observation) to embed</span>
<span class="n">n_samples_range</span> <span class="o">=</span> <a href="http://docs.scipy.org/doc/numpy-1.8.1/reference/generated/numpy.logspace.html#numpy.logspace" title="View documentation for numpy.logspace"><span class="n">np</span><span class="o">.</span><span class="n">logspace</span></a><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">9</span><span class="p">,</span> <span class="mi">9</span><span class="p">)</span>
<a href="http://matplotlib.org/api/_as_gen/matplotlib.figure.AxesStack.html#matplotlib.figure" title="View documentation for matplotlib.pyplot.figure"><span class="n">plt</span><span class="o">.</span><span class="n">figure</span></a><span class="p">()</span>
<span class="k">for</span> <span class="n">eps</span><span class="p">,</span> <span class="n">color</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">eps_range</span><span class="p">,</span> <span class="n">colors</span><span class="p">):</span>
<span class="n">min_n_components</span> <span class="o">=</span> <a href="../modules/generated/sklearn.random_projection.johnson_lindenstrauss_min_dim.html#sklearn.random_projection.johnson_lindenstrauss_min_dim" title="View documentation for sklearn.random_projection.johnson_lindenstrauss_min_dim"><span class="n">johnson_lindenstrauss_min_dim</span></a><span class="p">(</span><span class="n">n_samples_range</span><span class="p">,</span> <span class="n">eps</span><span class="o">=</span><span class="n">eps</span><span class="p">)</span>
<a href="http://matplotlib.org/api/_as_gen/matplotlib.pyplot.loglog.html#matplotlib.pyplot.loglog" title="View documentation for matplotlib.pyplot.loglog"><span class="n">plt</span><span class="o">.</span><span class="n">loglog</span></a><span class="p">(</span><span class="n">n_samples_range</span><span class="p">,</span> <span class="n">min_n_components</span><span class="p">,</span> <span class="n">color</span><span class="o">=</span><span class="n">color</span><span class="p">)</span>
<a href="http://matplotlib.org/api/legend_api.html#matplotlib.legend" title="View documentation for matplotlib.pyplot.legend"><span class="n">plt</span><span class="o">.</span><span class="n">legend</span></a><span class="p">([</span><span class="s2">"eps = </span><span class="si">%0.1f</span><span class="s2">"</span> <span class="o">%</span> <span class="n">eps</span> <span class="k">for</span> <span class="n">eps</span> <span class="ow">in</span> <span class="n">eps_range</span><span class="p">],</span> <span class="n">loc</span><span class="o">=</span><span class="s2">"lower right"</span><span class="p">)</span>
<a href="http://matplotlib.org/api/_as_gen/matplotlib.pyplot.xlabel.html#matplotlib.pyplot.xlabel" title="View documentation for matplotlib.pyplot.xlabel"><span class="n">plt</span><span class="o">.</span><span class="n">xlabel</span></a><span class="p">(</span><span class="s2">"Number of observations to eps-embed"</span><span class="p">)</span>
<a href="http://matplotlib.org/api/_as_gen/matplotlib.pyplot.ylabel.html#matplotlib.pyplot.ylabel" title="View documentation for matplotlib.pyplot.ylabel"><span class="n">plt</span><span class="o">.</span><span class="n">ylabel</span></a><span class="p">(</span><span class="s2">"Minimum number of dimensions"</span><span class="p">)</span>
<a href="http://matplotlib.org/api/_as_gen/matplotlib.pyplot.title.html#matplotlib.pyplot.title" title="View documentation for matplotlib.pyplot.title"><span class="n">plt</span><span class="o">.</span><span class="n">title</span></a><span class="p">(</span><span class="s2">"Johnson-Lindenstrauss bounds:</span><span class="se">\n</span><span class="s2">n_samples vs n_components"</span><span class="p">)</span>
<span class="c1"># range of admissible distortions</span>
<span class="n">eps_range</span> <span class="o">=</span> <a href="http://docs.scipy.org/doc/numpy-1.8.1/reference/generated/numpy.linspace.html#numpy.linspace" title="View documentation for numpy.linspace"><span class="n">np</span><span class="o">.</span><span class="n">linspace</span></a><span class="p">(</span><span class="mf">0.01</span><span class="p">,</span> <span class="mf">0.99</span><span class="p">,</span> <span class="mi">100</span><span class="p">)</span>
<span class="c1"># range of number of samples (observation) to embed</span>
<span class="n">n_samples_range</span> <span class="o">=</span> <a href="http://docs.scipy.org/doc/numpy-1.8.1/reference/generated/numpy.logspace.html#numpy.logspace" title="View documentation for numpy.logspace"><span class="n">np</span><span class="o">.</span><span class="n">logspace</span></a><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">6</span><span class="p">,</span> <span class="mi">5</span><span class="p">)</span>
<span class="n">colors</span> <span class="o">=</span> <span class="n">plt</span><span class="o">.</span><span class="n">cm</span><span class="o">.</span><span class="n">Blues</span><span class="p">(</span><a href="http://docs.scipy.org/doc/numpy-1.8.1/reference/generated/numpy.linspace.html#numpy.linspace" title="View documentation for numpy.linspace"><span class="n">np</span><span class="o">.</span><span class="n">linspace</span></a><span class="p">(</span><span class="mf">0.3</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">n_samples_range</span><span class="p">)))</span>
<a href="http://matplotlib.org/api/_as_gen/matplotlib.figure.AxesStack.html#matplotlib.figure" title="View documentation for matplotlib.pyplot.figure"><span class="n">plt</span><span class="o">.</span><span class="n">figure</span></a><span class="p">()</span>
<span class="k">for</span> <span class="n">n_samples</span><span class="p">,</span> <span class="n">color</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">n_samples_range</span><span class="p">,</span> <span class="n">colors</span><span class="p">):</span>
<span class="n">min_n_components</span> <span class="o">=</span> <a href="../modules/generated/sklearn.random_projection.johnson_lindenstrauss_min_dim.html#sklearn.random_projection.johnson_lindenstrauss_min_dim" title="View documentation for sklearn.random_projection.johnson_lindenstrauss_min_dim"><span class="n">johnson_lindenstrauss_min_dim</span></a><span class="p">(</span><span class="n">n_samples</span><span class="p">,</span> <span class="n">eps</span><span class="o">=</span><span class="n">eps_range</span><span class="p">)</span>
<a href="http://matplotlib.org/api/_as_gen/matplotlib.pyplot.semilogy.html#matplotlib.pyplot.semilogy" title="View documentation for matplotlib.pyplot.semilogy"><span class="n">plt</span><span class="o">.</span><span class="n">semilogy</span></a><span class="p">(</span><span class="n">eps_range</span><span class="p">,</span> <span class="n">min_n_components</span><span class="p">,</span> <span class="n">color</span><span class="o">=</span><span class="n">color</span><span class="p">)</span>
<a href="http://matplotlib.org/api/legend_api.html#matplotlib.legend" title="View documentation for matplotlib.pyplot.legend"><span class="n">plt</span><span class="o">.</span><span class="n">legend</span></a><span class="p">([</span><span class="s2">"n_samples = </span><span class="si">%d</span><span class="s2">"</span> <span class="o">%</span> <span class="n">n</span> <span class="k">for</span> <span class="n">n</span> <span class="ow">in</span> <span class="n">n_samples_range</span><span class="p">],</span> <span class="n">loc</span><span class="o">=</span><span class="s2">"upper right"</span><span class="p">)</span>
<a href="http://matplotlib.org/api/_as_gen/matplotlib.pyplot.xlabel.html#matplotlib.pyplot.xlabel" title="View documentation for matplotlib.pyplot.xlabel"><span class="n">plt</span><span class="o">.</span><span class="n">xlabel</span></a><span class="p">(</span><span class="s2">"Distortion eps"</span><span class="p">)</span>
<a href="http://matplotlib.org/api/_as_gen/matplotlib.pyplot.ylabel.html#matplotlib.pyplot.ylabel" title="View documentation for matplotlib.pyplot.ylabel"><span class="n">plt</span><span class="o">.</span><span class="n">ylabel</span></a><span class="p">(</span><span class="s2">"Minimum number of dimensions"</span><span class="p">)</span>
<a href="http://matplotlib.org/api/_as_gen/matplotlib.pyplot.title.html#matplotlib.pyplot.title" title="View documentation for matplotlib.pyplot.title"><span class="n">plt</span><span class="o">.</span><span class="n">title</span></a><span class="p">(</span><span class="s2">"Johnson-Lindenstrauss bounds:</span><span class="se">\n</span><span class="s2">n_components vs eps"</span><span class="p">)</span>
<span class="c1"># Part 2: perform sparse random projection of some digits images which are</span>
<span class="c1"># quite low dimensional and dense or documents of the 20 newsgroups dataset</span>
<span class="c1"># which is both high dimensional and sparse</span>
<span class="k">if</span> <span class="s1">'--twenty-newsgroups'</span> <span class="ow">in</span> <span class="n">sys</span><span class="o">.</span><span class="n">argv</span><span class="p">:</span>
<span class="c1"># Need an internet connection hence not enabled by default</span>
<span class="n">data</span> <span class="o">=</span> <a href="../modules/generated/sklearn.datasets.fetch_20newsgroups_vectorized.html#sklearn.datasets.fetch_20newsgroups_vectorized" title="View documentation for sklearn.datasets.fetch_20newsgroups_vectorized"><span class="n">fetch_20newsgroups_vectorized</span></a><span class="p">()</span><span class="o">.</span><span class="n">data</span><span class="p">[:</span><span class="mi">500</span><span class="p">]</span>
<span class="k">else</span><span class="p">:</span>
<span class="n">data</span> <span class="o">=</span> <a href="../modules/generated/sklearn.datasets.load_digits.html#sklearn.datasets.load_digits" title="View documentation for sklearn.datasets.load_digits"><span class="n">load_digits</span></a><span class="p">()</span><span class="o">.</span><span class="n">data</span><span class="p">[:</span><span class="mi">500</span><span class="p">]</span>
<span class="n">n_samples</span><span class="p">,</span> <span class="n">n_features</span> <span class="o">=</span> <span class="n">data</span><span class="o">.</span><span class="n">shape</span>
<span class="k">print</span><span class="p">(</span><span class="s2">"Embedding </span><span class="si">%d</span><span class="s2"> samples with dim </span><span class="si">%d</span><span class="s2"> using various random projections"</span>
<span class="o">%</span> <span class="p">(</span><span class="n">n_samples</span><span class="p">,</span> <span class="n">n_features</span><span class="p">))</span>
<span class="n">n_components_range</span> <span class="o">=</span> <a href="http://docs.scipy.org/doc/numpy-1.8.1/reference/generated/numpy.array.html#numpy.array" title="View documentation for numpy.array"><span class="n">np</span><span class="o">.</span><span class="n">array</span></a><span class="p">([</span><span class="mi">300</span><span class="p">,</span> <span class="mi">1000</span><span class="p">,</span> <span class="mi">10000</span><span class="p">])</span>
<span class="n">dists</span> <span class="o">=</span> <span class="n">euclidean_distances</span><span class="p">(</span><span class="n">data</span><span class="p">,</span> <span class="n">squared</span><span class="o">=</span><span class="bp">True</span><span class="p">)</span><span class="o">.</span><span class="n">ravel</span><span class="p">()</span>
<span class="c1"># select only non-identical samples pairs</span>
<span class="n">nonzero</span> <span class="o">=</span> <span class="n">dists</span> <span class="o">!=</span> <span class="mi">0</span>
<span class="n">dists</span> <span class="o">=</span> <span class="n">dists</span><span class="p">[</span><span class="n">nonzero</span><span class="p">]</span>
<span class="k">for</span> <span class="n">n_components</span> <span class="ow">in</span> <span class="n">n_components_range</span><span class="p">:</span>
<span class="n">t0</span> <span class="o">=</span> <span class="n">time</span><span class="p">()</span>
<span class="n">rp</span> <span class="o">=</span> <a href="../modules/generated/sklearn.random_projection.SparseRandomProjection.html#sklearn.random_projection.SparseRandomProjection" title="View documentation for sklearn.random_projection.SparseRandomProjection"><span class="n">SparseRandomProjection</span></a><span class="p">(</span><span class="n">n_components</span><span class="o">=</span><span class="n">n_components</span><span class="p">)</span>
<span class="n">projected_data</span> <span class="o">=</span> <span class="n">rp</span><span class="o">.</span><span class="n">fit_transform</span><span class="p">(</span><span class="n">data</span><span class="p">)</span>
<span class="k">print</span><span class="p">(</span><span class="s2">"Projected </span><span class="si">%d</span><span class="s2"> samples from </span><span class="si">%d</span><span class="s2"> to </span><span class="si">%d</span><span class="s2"> in </span><span class="si">%0.3f</span><span class="s2">s"</span>
<span class="o">%</span> <span class="p">(</span><span class="n">n_samples</span><span class="p">,</span> <span class="n">n_features</span><span class="p">,</span> <span class="n">n_components</span><span class="p">,</span> <span class="n">time</span><span class="p">()</span> <span class="o">-</span> <span class="n">t0</span><span class="p">))</span>
<span class="k">if</span> <span class="nb">hasattr</span><span class="p">(</span><span class="n">rp</span><span class="p">,</span> <span class="s1">'components_'</span><span class="p">):</span>
<span class="n">n_bytes</span> <span class="o">=</span> <span class="n">rp</span><span class="o">.</span><span class="n">components_</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">nbytes</span>
<span class="n">n_bytes</span> <span class="o">+=</span> <span class="n">rp</span><span class="o">.</span><span class="n">components_</span><span class="o">.</span><span class="n">indices</span><span class="o">.</span><span class="n">nbytes</span>
<span class="k">print</span><span class="p">(</span><span class="s2">"Random matrix with size: </span><span class="si">%0.3f</span><span class="s2">MB"</span> <span class="o">%</span> <span class="p">(</span><span class="n">n_bytes</span> <span class="o">/</span> <span class="mf">1e6</span><span class="p">))</span>
<span class="n">projected_dists</span> <span class="o">=</span> <span class="n">euclidean_distances</span><span class="p">(</span>
<span class="n">projected_data</span><span class="p">,</span> <span class="n">squared</span><span class="o">=</span><span class="bp">True</span><span class="p">)</span><span class="o">.</span><span class="n">ravel</span><span class="p">()[</span><span class="n">nonzero</span><span class="p">]</span>
<a href="http://matplotlib.org/api/_as_gen/matplotlib.figure.AxesStack.html#matplotlib.figure" title="View documentation for matplotlib.pyplot.figure"><span class="n">plt</span><span class="o">.</span><span class="n">figure</span></a><span class="p">()</span>
<a href="http://matplotlib.org/api/_as_gen/matplotlib.pyplot.hexbin.html#matplotlib.pyplot.hexbin" title="View documentation for matplotlib.pyplot.hexbin"><span class="n">plt</span><span class="o">.</span><span class="n">hexbin</span></a><span class="p">(</span><span class="n">dists</span><span class="p">,</span> <span class="n">projected_dists</span><span class="p">,</span> <span class="n">gridsize</span><span class="o">=</span><span class="mi">100</span><span class="p">,</span> <span class="n">cmap</span><span class="o">=</span><span class="n">plt</span><span class="o">.</span><span class="n">cm</span><span class="o">.</span><span class="n">PuBu</span><span class="p">)</span>
<a href="http://matplotlib.org/api/_as_gen/matplotlib.pyplot.xlabel.html#matplotlib.pyplot.xlabel" title="View documentation for matplotlib.pyplot.xlabel"><span class="n">plt</span><span class="o">.</span><span class="n">xlabel</span></a><span class="p">(</span><span class="s2">"Pairwise squared distances in original space"</span><span class="p">)</span>
<a href="http://matplotlib.org/api/_as_gen/matplotlib.pyplot.ylabel.html#matplotlib.pyplot.ylabel" title="View documentation for matplotlib.pyplot.ylabel"><span class="n">plt</span><span class="o">.</span><span class="n">ylabel</span></a><span class="p">(</span><span class="s2">"Pairwise squared distances in projected space"</span><span class="p">)</span>
<a href="http://matplotlib.org/api/_as_gen/matplotlib.pyplot.title.html#matplotlib.pyplot.title" title="View documentation for matplotlib.pyplot.title"><span class="n">plt</span><span class="o">.</span><span class="n">title</span></a><span class="p">(</span><span class="s2">"Pairwise distances distribution for n_components=</span><span class="si">%d</span><span class="s2">"</span> <span class="o">%</span>
<span class="n">n_components</span><span class="p">)</span>
<span class="n">cb</span> <span class="o">=</span> <a href="http://matplotlib.org/api/colorbar_api.html#matplotlib.colorbar" title="View documentation for matplotlib.pyplot.colorbar"><span class="n">plt</span><span class="o">.</span><span class="n">colorbar</span></a><span class="p">()</span>
<span class="n">cb</span><span class="o">.</span><span class="n">set_label</span><span class="p">(</span><span class="s1">'Sample pairs counts'</span><span class="p">)</span>
<span class="n">rates</span> <span class="o">=</span> <span class="n">projected_dists</span> <span class="o">/</span> <span class="n">dists</span>
<span class="k">print</span><span class="p">(</span><span class="s2">"Mean distances rate: </span><span class="si">%0.2f</span><span class="s2"> (</span><span class="si">%0.2f</span><span class="s2">)"</span>
<span class="o">%</span> <span class="p">(</span><a href="http://docs.scipy.org/doc/numpy-1.8.1/reference/generated/numpy.mean.html#numpy.mean" title="View documentation for numpy.mean"><span class="n">np</span><span class="o">.</span><span class="n">mean</span></a><span class="p">(</span><span class="n">rates</span><span class="p">),</span> <a href="http://docs.scipy.org/doc/numpy-1.8.1/reference/generated/numpy.std.html#numpy.std" title="View documentation for numpy.std"><span class="n">np</span><span class="o">.</span><span class="n">std</span></a><span class="p">(</span><span class="n">rates</span><span class="p">)))</span>
<a href="http://matplotlib.org/api/_as_gen/matplotlib.figure.AxesStack.html#matplotlib.figure" title="View documentation for matplotlib.pyplot.figure"><span class="n">plt</span><span class="o">.</span><span class="n">figure</span></a><span class="p">()</span>
<a href="http://matplotlib.org/api/_as_gen/matplotlib.pyplot.hist.html#matplotlib.pyplot.hist" title="View documentation for matplotlib.pyplot.hist"><span class="n">plt</span><span class="o">.</span><span class="n">hist</span></a><span class="p">(</span><span class="n">rates</span><span class="p">,</span> <span class="n">bins</span><span class="o">=</span><span class="mi">50</span><span class="p">,</span> <span class="n">normed</span><span class="o">=</span><span class="bp">True</span><span class="p">,</span> <span class="nb">range</span><span class="o">=</span><span class="p">(</span><span class="mf">0.</span><span class="p">,</span> <span class="mf">2.</span><span class="p">),</span> <span class="n">edgecolor</span><span class="o">=</span><span class="s1">'k'</span><span class="p">)</span>
<a href="http://matplotlib.org/api/_as_gen/matplotlib.pyplot.xlabel.html#matplotlib.pyplot.xlabel" title="View documentation for matplotlib.pyplot.xlabel"><span class="n">plt</span><span class="o">.</span><span class="n">xlabel</span></a><span class="p">(</span><span class="s2">"Squared distances rate: projected / original"</span><span class="p">)</span>
<a href="http://matplotlib.org/api/_as_gen/matplotlib.pyplot.ylabel.html#matplotlib.pyplot.ylabel" title="View documentation for matplotlib.pyplot.ylabel"><span class="n">plt</span><span class="o">.</span><span class="n">ylabel</span></a><span class="p">(</span><span class="s2">"Distribution of samples pairs"</span><span class="p">)</span>
<a href="http://matplotlib.org/api/_as_gen/matplotlib.pyplot.title.html#matplotlib.pyplot.title" title="View documentation for matplotlib.pyplot.title"><span class="n">plt</span><span class="o">.</span><span class="n">title</span></a><span class="p">(</span><span class="s2">"Histogram of pairwise distance rates for n_components=</span><span class="si">%d</span><span class="s2">"</span> <span class="o">%</span>
<span class="n">n_components</span><span class="p">)</span>
<span class="c1"># TODO: compute the expected value of eps and add them to the previous plot</span>
<span class="c1"># as vertical lines / region</span>
<a href="http://matplotlib.org/api/_as_gen/matplotlib.pyplot.show.html#matplotlib.pyplot.show" title="View documentation for matplotlib.pyplot.show"><span class="n">plt</span><span class="o">.</span><span class="n">show</span></a><span class="p">()</span>
</pre></div>
</div>
<p><strong>Total running time of the script:</strong> ( 0 minutes 2.190 seconds)</p>
<div class="sphx-glr-footer docutils container">
<div class="sphx-glr-download docutils container">
<a class="reference download internal" href="../_downloads/plot_johnson_lindenstrauss_bound.py" download=""><code class="xref download docutils literal"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">plot_johnson_lindenstrauss_bound.py</span></code></a></div>
<div class="sphx-glr-download docutils container">
<a class="reference download internal" href="../_downloads/plot_johnson_lindenstrauss_bound.ipynb" download=""><code class="xref download docutils literal"><span class="pre">Download</span> <span class="pre">Jupyter</span> <span class="pre">notebook:</span> <span class="pre">plot_johnson_lindenstrauss_bound.ipynb</span></code></a></div>
</div>
<p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.readthedocs.io">Generated by Sphinx-Gallery</a></p>
</div>
</div>
</div>
</div>
</div>
<div class="clearer"></div>
</div>
</div>
<div class="footer">
© 2007 - 2017, scikit-learn developers (BSD License).
<a href="../_sources/auto_examples/plot_johnson_lindenstrauss_bound.rst.txt" rel="nofollow">Show this page source</a>
</div>
<div class="rel">
<div class="buttonPrevious">
<a href="plot_multilabel.html">Previous
</a>
</div>
<div class="buttonNext">
<a href="plot_kernel_ridge_regression.html">Next
</a>
</div>
</div>
<script type="text/javascript">
var _gaq = _gaq || [];
_gaq.push(['_setAccount', 'UA-22606712-2']);
_gaq.push(['_trackPageview']);
(function() {
var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true;
ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js';
var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s);
})();
</script>
<script>
(function() {
var cx = '016639176250731907682:tjtqbvtvij0';
var gcse = document.createElement('script'); gcse.type = 'text/javascript'; gcse.async = true;
gcse.src = 'https://cse.google.com/cse.js?cx=' + cx;
var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(gcse, s);
})();
</script>
<script src="https://scikit-learn.org/versionwarning.js"></script>
</body>
</html>