-
Notifications
You must be signed in to change notification settings - Fork 11
/
index.html
836 lines (834 loc) · 432 KB
/
index.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
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
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
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
<!DOCTYPE html>
<html>
<head>
<link crossorigin="anonymous" href="https://maxcdn.bootstrapcdn.com/bootstrap/4.0.0/css/bootstrap.min.css" integrity="sha384-Gn5384xqQ1aoWXA+058RXPxPg6fy4IWvTNh0E263XmFcJlSAwiGgFAW/dAiS6JXm" rel="stylesheet"/>
<link href="css/notebook.css" rel="stylesheet"/>
<link href="css/pygments/notebook/colorful.css" rel="stylesheet"/>
<link href="css/main.css" rel="stylesheet"/>
</head>
<body>
<div class="container">
<h1>
Integrating a Jupyter Notebook in a Web Page
</h1>
<p>
Some text.
</p>
<h2>
Maths
</h2>
<p>
Here is a simple inline equation: $E=mc^2$.
</p>
<p>
And for longer equations, use the standard math mode:
</p>
$$
\mathcal{L}=\bar{\psi}\left(i\gamma^{\mu}\partial_ {\mu}-m\right)\psi
$$
<h2>
A code section
</h2>
<p>
Here is some code:
</p>
<div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">foo</span>
<span class="k">print</span> <span class="n">foo</span><span class="o">.</span><span class="n">bar</span><span class="p">()</span>
</pre></div>
<div class="cell border-box-sizing text_cell rendered">
<div class="prompt input_prompt">
</div>
<div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<h2 id="Introduction">
Introduction
<a class="anchor-link" href="#Introduction">
¶
</a>
</h2>
<p>
Overfitting is one of the most important issues in machine learning, if not the most important.
</p>
<p>
In this post, I'll illustrate overfitting in the context of a small 2D classification problem.
</p>
<p>
But what is going to be explained here is important, and should be kept in mind at all times when working on more complex problems.
</p>
<p>
You will learn:
</p>
<ul>
<li>
what is overfitting, and see it with your own eyes
</li>
<li>
how to avoid overfitting
</li>
<li>
that if you use a neural network that is too complex for the amount of data you have, you'll just get crap.
</li>
<li>
that complex models are still necessary to deal with complex problems, otherwise you get underfitting.
</li>
</ul>
<p>
<strong>
Prerequisites
</strong>
:
</p>
<ul>
<li>
You should know a bit of numpy and matplotlib, and be familiar with classification in 2D. If not, you can have a look at
<a href="https://thedatafrog.com/logistic-regression-neural-network/">
my tutorial about logistic regression and neural networks for 2D classification
</a>
</li>
</ul>
<p>
To run the code in this tutorial, you can simply
<a href="https://colab.research.google.com/github/cbernet/maldives/blob/master/overfitting/overfitting.ipynb">
open it in google colab
</a>
.
</p>
<p>
Alternatively, if you have
<a href="https://www.anaconda.com">
Anaconda
</a>
installed (2.X or 3.X), you can:
</p>
<ul>
<li>
<a href="https://github.com/cbernet/maldives/archive/master.zip">
download the repository containing this notebook
</a>
</li>
<li>
unzip it, say to
<code>
Downloads/maldives-master
</code>
</li>
<li>
launch a jupyter notebook from the anaconda navigator
</li>
<li>
in jupyter notebook, navigate to
<code>
Downloads/maldives-master/overfitting
</code>
</li>
<li>
open
<code>
overfitting.ipynb
</code>
</li>
</ul>
<p>
First, let's setup our tools:
</p>
</div>
</div>
</div>
<div class="cell border-box-sizing code_cell rendered">
<div class="input">
<div class="prompt input_prompt">
In [1]:
</div>
<div class="inner_cell">
<div class="input_area">
<div class=" highlight hl-ipython3">
<pre><span></span><span class="kn">import</span> <span class="nn">matplotlib.pyplot</span> <span class="k">as</span> <span class="nn">plt</span>
<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
<span class="o">%</span><span class="k">matplotlib</span> inline
<span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">seed</span><span class="p">(</span><span class="mh">0xdeadbeef</span><span class="p">)</span>
<span class="c1"># blah</span>
</pre>
</div>
</div>
</div>
</div>
</div>
<div class="cell border-box-sizing text_cell rendered">
<div class="prompt input_prompt">
</div>
<div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<h2 id="Building-a-small-dataset">
Building a small dataset
<a class="anchor-link" href="#Building-a-small-dataset">
¶
</a>
</h2>
<p>
Let's create a sample of examples with two values x1 and x2, with two categories.
For category 0, the underlying probability distribution is a 2D Gaussian centered on (0,0), with width = 1 along both directions. For category 1, the Gaussian is centered on (1,1). We assign label 0 to category 0, and label 1 to category 1.
</p>
</div>
</div>
</div>
<div class="cell border-box-sizing code_cell rendered">
<div class="input">
<div class="prompt input_prompt">
In [17]:
</div>
<div class="inner_cell">
<div class="input_area">
<div class=" highlight hl-ipython3">
<pre><span></span><span class="k">def</span> <span class="nf">make_sample</span><span class="p">(</span><span class="n">nexamples</span><span class="p">,</span> <span class="n">means</span><span class="o">=</span><span class="p">([</span><span class="mf">0.</span><span class="p">,</span><span class="mf">0.</span><span class="p">],[</span><span class="mf">1.</span><span class="p">,</span><span class="mf">1.</span><span class="p">]),</span> <span class="n">sigma</span><span class="o">=</span><span class="mf">1.</span><span class="p">):</span>
<span class="n">normal</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">multivariate_normal</span>
<span class="c1"># squared width:</span>
<span class="n">s2</span> <span class="o">=</span> <span class="n">sigma</span><span class="o">**</span><span class="mf">2.</span>
<span class="c1"># below, we provide the coordinates of the mean as </span>
<span class="c1"># a first argument, and then the covariance matrix</span>
<span class="c1"># which describes the width of the Gaussian along the </span>
<span class="c1"># two directions. </span>
<span class="c1"># we generate nexamples examples for each category</span>
<span class="n">sgx0</span> <span class="o">=</span> <span class="n">normal</span><span class="p">(</span><span class="n">means</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="p">[[</span><span class="n">s2</span><span class="p">,</span> <span class="mf">0.</span><span class="p">],</span> <span class="p">[</span><span class="mf">0.</span><span class="p">,</span><span class="n">s2</span><span class="p">]],</span> <span class="n">nexamples</span><span class="p">)</span>
<span class="n">sgx1</span> <span class="o">=</span> <span class="n">normal</span><span class="p">(</span><span class="n">means</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="p">[[</span><span class="n">s2</span><span class="p">,</span> <span class="mf">0.</span><span class="p">],</span> <span class="p">[</span><span class="mf">0.</span><span class="p">,</span><span class="n">s2</span><span class="p">]],</span> <span class="n">nexamples</span><span class="p">)</span>
<span class="c1"># setting the labels for each category</span>
<span class="n">sgy0</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">zeros</span><span class="p">((</span><span class="n">nexamples</span><span class="p">,))</span>
<span class="n">sgy1</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">ones</span><span class="p">((</span><span class="n">nexamples</span><span class="p">,))</span>
<span class="n">sgx</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">concatenate</span><span class="p">([</span><span class="n">sgx0</span><span class="p">,</span><span class="n">sgx1</span><span class="p">])</span>
<span class="n">sgy</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">concatenate</span><span class="p">([</span><span class="n">sgy0</span><span class="p">,</span><span class="n">sgy1</span><span class="p">])</span>
<span class="k">return</span> <span class="n">sgx</span><span class="p">,</span> <span class="n">sgy</span>
</pre>
</div>
</div>
</div>
</div>
</div>
<div class="cell border-box-sizing text_cell rendered">
<div class="prompt input_prompt">
</div>
<div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<p>
Here, we create a very small training sample with only 30 examples per category, and a test sample with 200 examples per category. We're using such a small training sample because, as will be shown in this post, small samples are very easy to overfit.
</p>
</div>
</div>
</div>
<div class="cell border-box-sizing code_cell rendered">
<div class="input">
<div class="prompt input_prompt">
In [18]:
</div>
<div class="inner_cell">
<div class="input_area">
<div class=" highlight hl-ipython3">
<pre><span></span><span class="n">sgx</span><span class="p">,</span> <span class="n">sgy</span> <span class="o">=</span> <span class="n">make_sample</span><span class="p">(</span><span class="mi">30</span><span class="p">)</span>
<span class="n">tgx</span><span class="p">,</span> <span class="n">tgy</span> <span class="o">=</span> <span class="n">make_sample</span><span class="p">(</span><span class="mi">200</span><span class="p">)</span>
</pre>
</div>
</div>
</div>
</div>
</div>
<div class="cell border-box-sizing code_cell rendered">
<div class="input">
<div class="prompt input_prompt">
In [4]:
</div>
<div class="inner_cell">
<div class="input_area">
<div class=" highlight hl-ipython3">
<pre><span></span><span class="c1"># note how the two categories are plotted</span>
<span class="c1"># together in one go by providing the </span>
<span class="c1"># label array as color argument (c=sgy)</span>
<span class="n">plt</span><span class="o">.</span><span class="n">scatter</span><span class="p">(</span><span class="n">sgx</span><span class="p">[:,</span><span class="mi">0</span><span class="p">],</span> <span class="n">sgx</span><span class="p">[:,</span><span class="mi">1</span><span class="p">],</span> <span class="n">alpha</span><span class="o">=</span><span class="mf">0.5</span><span class="p">,</span> <span class="n">c</span><span class="o">=</span><span class="n">sgy</span><span class="p">)</span>
<span class="n">plt</span><span class="o">.</span><span class="n">xlabel</span><span class="p">(</span><span class="s1">'x1'</span><span class="p">)</span>
<span class="n">plt</span><span class="o">.</span><span class="n">ylabel</span><span class="p">(</span><span class="s1">'x2'</span><span class="p">)</span>
</pre>
</div>
</div>
</div>
</div>
<div class="output_wrapper">
<div class="output">
<div class="output_area">
<div class="prompt output_prompt">
Out[4]:
</div>
<div class="output_text output_subarea output_execute_result">
<pre>Text(0, 0.5, 'x2')</pre>
</div>
</div>
<div class="output_area">
<div class="prompt">
</div>
<div class="output_png output_subarea ">
<img src="data:image/png;base64,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
"/>
</div>
</div>
</div>
</div>
</div>
<div class="cell border-box-sizing text_cell rendered">
<div class="prompt input_prompt">
</div>
<div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<p>
We see that with such a low number of examples, it is not obvious that the samples follow Gaussian probability density functions. Also, since the Gaussians are so close, it's going to be quite difficult to separate the two categories.
</p>
<h2 id="Overfitting">
Overfitting
<a class="anchor-link" href="#Overfitting">
¶
</a>
</h2>
<p>
Let's try anyway with a neural network from scikit-learn. Here is an explanation of the parameters I use below:
</p>
<ul>
<li>
three hidden layers with 50 neurons each. I've chosen this fairly complex configuration on purpose to illustrate overfitting, which occurs when the model is too complex for the amount of data in the training sample.
</li>
<li>
relu activation, because relu makes the training easier in neural nets with hidden layers.
</li>
<li>
an increased maximum number of iterations, so that the network has time to converge
</li>
<li>
a fixed random seed so that you can get the exact same results as me, every time you run the code
</li>
</ul>
</div>
</div>
</div>
<div class="cell border-box-sizing code_cell rendered">
<div class="input">
<div class="prompt input_prompt">
In [5]:
</div>
<div class="inner_cell">
<div class="input_area">
<div class=" highlight hl-ipython3">
<pre><span></span><span class="kn">from</span> <span class="nn">sklearn.neural_network</span> <span class="k">import</span> <span class="n">MLPClassifier</span>
<span class="n">mlp</span> <span class="o">=</span> <span class="n">MLPClassifier</span><span class="p">(</span><span class="n">hidden_layer_sizes</span><span class="o">=</span><span class="p">(</span><span class="mi">50</span><span class="p">,</span><span class="mi">50</span><span class="p">,</span><span class="mi">50</span><span class="p">),</span> <span class="n">activation</span><span class="o">=</span><span class="s1">'relu'</span><span class="p">,</span> <span class="n">max_iter</span><span class="o">=</span><span class="mi">10000</span><span class="p">,</span> <span class="n">random_state</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
<span class="n">mlp</span><span class="o">.</span><span class="n">fit</span><span class="p">(</span><span class="n">sgx</span><span class="p">,</span><span class="n">sgy</span><span class="p">)</span>
</pre>
</div>
</div>
</div>
</div>
<div class="output_wrapper">
<div class="output">
<div class="output_area">
<div class="prompt output_prompt">
Out[5]:
</div>
<div class="output_text output_subarea output_execute_result">
<pre>MLPClassifier(activation='relu', alpha=0.0001, batch_size='auto', beta_1=0.9,
beta_2=0.999, early_stopping=False, epsilon=1e-08,
hidden_layer_sizes=(50, 50, 50), learning_rate='constant',
learning_rate_init=0.001, max_iter=10000, momentum=0.9,
n_iter_no_change=10, nesterovs_momentum=True, power_t=0.5,
random_state=1, shuffle=True, solver='adam', tol=0.0001,
validation_fraction=0.1, verbose=False, warm_start=False)</pre>
</div>
</div>
</div>
</div>
</div>
<div class="cell border-box-sizing text_cell rendered">
<div class="prompt input_prompt">
</div>
<div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<p>
Now let's define a small function to plot our results. The function will plot the examples in the two categories, as well as the probability that an (x1,x2) point belongs to category 1 (black means that this probability is close to 1, and white to 0.)
</p>
</div>
</div>
</div>
<div class="cell border-box-sizing code_cell rendered">
<div class="input">
<div class="prompt input_prompt">
In [6]:
</div>
<div class="inner_cell">
<div class="input_area">
<div class=" highlight hl-ipython3">
<pre><span></span><span class="k">def</span> <span class="nf">plot_result</span><span class="p">(</span><span class="n">sample</span><span class="p">,</span> <span class="n">targets</span><span class="p">,</span> <span class="n">linrange</span><span class="o">=</span><span class="p">(</span><span class="o">-</span><span class="mi">5</span><span class="p">,</span><span class="mi">5</span><span class="p">,</span><span class="mi">101</span><span class="p">)):</span>
<span class="n">xmin</span><span class="p">,</span> <span class="n">xmax</span><span class="p">,</span> <span class="n">npoints</span> <span class="o">=</span> <span class="n">linrange</span>
<span class="n">gridx1</span><span class="p">,</span> <span class="n">gridx2</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">meshgrid</span><span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="n">xmin</span><span class="p">,</span><span class="n">xmax</span><span class="p">,</span><span class="n">npoints</span><span class="p">),</span> <span class="n">np</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="n">xmin</span><span class="p">,</span><span class="n">xmax</span><span class="p">,</span><span class="n">npoints</span><span class="p">))</span>
<span class="n">grid</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">c_</span><span class="p">[</span><span class="n">gridx1</span><span class="o">.</span><span class="n">flatten</span><span class="p">(),</span> <span class="n">gridx2</span><span class="o">.</span><span class="n">flatten</span><span class="p">()]</span>
<span class="n">probs</span> <span class="o">=</span> <span class="n">mlp</span><span class="o">.</span><span class="n">predict_proba</span><span class="p">(</span><span class="n">grid</span><span class="p">)</span>
<span class="n">plt</span><span class="o">.</span><span class="n">pcolor</span><span class="p">(</span><span class="n">gridx1</span><span class="p">,</span> <span class="n">gridx2</span><span class="p">,</span> <span class="n">probs</span><span class="p">[:,</span><span class="mi">1</span><span class="p">]</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">npoints</span><span class="p">,</span><span class="n">npoints</span><span class="p">),</span> <span class="n">cmap</span><span class="o">=</span><span class="s1">'binary'</span><span class="p">)</span>
<span class="n">plt</span><span class="o">.</span><span class="n">colorbar</span><span class="p">()</span>
<span class="n">plt</span><span class="o">.</span><span class="n">scatter</span><span class="p">(</span><span class="n">sample</span><span class="p">[:,</span><span class="mi">0</span><span class="p">],</span> <span class="n">sample</span><span class="p">[:,</span><span class="mi">1</span><span class="p">],</span> <span class="n">c</span><span class="o">=</span><span class="n">targets</span><span class="p">,</span> <span class="n">cmap</span><span class="o">=</span><span class="s1">'plasma'</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="mf">0.5</span><span class="p">,</span> <span class="n">marker</span><span class="o">=</span><span class="s1">'.'</span><span class="p">)</span>
<span class="n">plt</span><span class="o">.</span><span class="n">xlabel</span><span class="p">(</span><span class="s1">'x1'</span><span class="p">)</span>
<span class="n">plt</span><span class="o">.</span><span class="n">ylabel</span><span class="p">(</span><span class="s1">'x2'</span><span class="p">)</span>
<span class="n">plt</span><span class="o">.</span><span class="n">show</span><span class="p">()</span>
</pre>
</div>
</div>
</div>
</div>
</div>
<div class="cell border-box-sizing code_cell rendered">
<div class="input">
<div class="prompt input_prompt">
In [7]:
</div>
<div class="inner_cell">
<div class="input_area">
<div class=" highlight hl-ipython3">
<pre><span></span><span class="n">plot_result</span><span class="p">(</span><span class="n">sgx</span><span class="p">,</span><span class="n">sgy</span><span class="p">)</span>
</pre>
</div>
</div>
</div>
</div>
<div class="output_wrapper">
<div class="output">
<div class="output_area">
<div class="prompt">
</div>
<div class="output_png output_subarea ">
<img src="data:image/png;base64,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
"/>
</div>
</div>
</div>
</div>
</div>
<div class="cell border-box-sizing text_cell rendered">
<div class="prompt input_prompt">
</div>
<div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<p>
Very nice eagle, but the probability distribution is very far from being optimal. We see that at the frontier, the neural network does its best to follow the patterns of the training sample. It is able to do that because its large number of parameters make it very flexible and adaptive.
</p>
<p>
But let's see what happens if we plot the probability distribution with the larger test sample:
</p>
</div>
</div>
</div>
<div class="cell border-box-sizing code_cell rendered">
<div class="input">
<div class="prompt input_prompt">
In [8]:
</div>
<div class="inner_cell">
<div class="input_area">
<div class=" highlight hl-ipython3">
<pre><span></span><span class="n">plot_result</span><span class="p">(</span><span class="n">tgx</span><span class="p">,</span><span class="n">tgy</span><span class="p">)</span>
</pre>
</div>
</div>
</div>
</div>
<div class="output_wrapper">
<div class="output">
<div class="output_area">
<div class="prompt">
</div>
<div class="output_png output_subarea ">
<img src="data:image/png;base64,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
"/>
</div>
</div>
</div>
</div>
</div>
<div class="cell border-box-sizing text_cell rendered">
<div class="prompt input_prompt">
</div>
<div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<p>
There are many examples that are classified in the wrong category. This neural network is very good with the training sample, but it has lost its generality and is thus useless in practice.
</p>
<p>
<strong>
This is overfitting.
</strong>
</p>
<h2 id="Fixing-overfitting">
Fixing overfitting
<a class="anchor-link" href="#Fixing-overfitting">
¶
</a>
</h2>
<p>
Now let's try again, but with a much more simple network, with a single layer with five neurons. The network is trained with the small training sample, and displayed with the larger test sample:
</p>
</div>
</div>
</div>
<div class="cell border-box-sizing code_cell rendered">
<div class="input">
<div class="prompt input_prompt">
In [9]:
</div>
<div class="inner_cell">
<div class="input_area">
<div class=" highlight hl-ipython3">
<pre><span></span><span class="n">mlp</span> <span class="o">=</span> <span class="n">MLPClassifier</span><span class="p">(</span><span class="n">hidden_layer_sizes</span><span class="o">=</span><span class="p">(</span><span class="mi">5</span><span class="p">,),</span> <span class="n">activation</span><span class="o">=</span><span class="s1">'relu'</span><span class="p">,</span> <span class="n">max_iter</span><span class="o">=</span><span class="mi">10000</span><span class="p">,</span> <span class="n">random_state</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
<span class="n">mlp</span><span class="o">.</span><span class="n">fit</span><span class="p">(</span><span class="n">sgx</span><span class="p">,</span><span class="n">sgy</span><span class="p">)</span>
<span class="n">plot_result</span><span class="p">(</span><span class="n">tgx</span><span class="p">,</span><span class="n">tgy</span><span class="p">)</span>
</pre>
</div>
</div>
</div>
</div>
<div class="output_wrapper">
<div class="output">
<div class="output_area">
<div class="prompt">
</div>
<div class="output_png output_subarea ">
<img src="data:image/png;base64,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
"/>
</div>
</div>
</div>
</div>
</div>
<div class="cell border-box-sizing text_cell rendered">
<div class="prompt input_prompt">
</div>
<div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<p>
This time, overfitting is much less of an issue. The network does not have enough parameters to be able to follow the random patterns of the training sample. Therefore, it behaves quite well on the test sample.
</p>
<p>
Let's try something else. This time, we use the complex network, but we provide much more training data: 10,000 examples per category instead of 30.
</p>
</div>
</div>
</div>
<div class="cell border-box-sizing code_cell rendered">
<div class="input">
<div class="prompt input_prompt">
In [19]:
</div>
<div class="inner_cell">
<div class="input_area">
<div class=" highlight hl-ipython3">
<pre><span></span><span class="n">sgx</span><span class="p">,</span> <span class="n">sgy</span> <span class="o">=</span> <span class="n">make_sample</span><span class="p">(</span><span class="mi">10000</span><span class="p">)</span>
<span class="n">mlp</span> <span class="o">=</span> <span class="n">MLPClassifier</span><span class="p">(</span><span class="n">hidden_layer_sizes</span><span class="o">=</span><span class="p">(</span><span class="mi">50</span><span class="p">,</span><span class="mi">50</span><span class="p">,</span><span class="mi">50</span><span class="p">),</span> <span class="n">activation</span><span class="o">=</span><span class="s1">'relu'</span><span class="p">,</span> <span class="n">max_iter</span><span class="o">=</span><span class="mi">10000</span><span class="p">,</span> <span class="n">random_state</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
<span class="n">mlp</span><span class="o">.</span><span class="n">fit</span><span class="p">(</span><span class="n">sgx</span><span class="p">,</span><span class="n">sgy</span><span class="p">)</span>
<span class="n">plot_result</span><span class="p">(</span><span class="n">tgx</span><span class="p">,</span> <span class="n">tgy</span><span class="p">)</span>
</pre>
</div>
</div>
</div>
</div>
<div class="output_wrapper">
<div class="output">
<div class="output_area">
<div class="prompt">
</div>
<div class="output_png output_subarea ">
<img src="data:image/png;base64,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
"/>
</div>
</div>
</div>
</div>
</div>
<div class="cell border-box-sizing text_cell rendered">
<div class="prompt input_prompt">
</div>
<div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<p>
The network took a bit longer to train but this time, there is enough training data to properly constrain the parameters of the network, and the classification performance is going to be good in general.
</p>
</div>
</div>
</div>
<div class="cell border-box-sizing text_cell rendered">
<div class="prompt input_prompt">
</div>
<div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<h2 id="So-why-do-we-need-complex-networks-then?">
So why do we need complex networks then?
<a class="anchor-link" href="#So-why-do-we-need-complex-networks-then?">
¶
</a>
</h2>
<p>
Well, to describe complex data! And for these networks to be efficient in this task, we will need a lot of training data.
</p>
<p>
In this section, we'll build a complex dataset with a lot of data, and see how well we can classify it.
</p>
<p>
To build the dataset, we just reuse our make_sample function several times and concatenate the resulting samples:
</p>
</div>
</div>
</div>
<div class="cell border-box-sizing code_cell rendered">
<div class="input">
<div class="prompt input_prompt">
In [11]:
</div>
<div class="inner_cell">
<div class="input_area">
<div class=" highlight hl-ipython3">
<pre><span></span><span class="n">sgxa</span><span class="p">,</span> <span class="n">sgya</span> <span class="o">=</span> <span class="n">make_sample</span><span class="p">(</span><span class="mi">1000</span><span class="p">,</span> <span class="p">([</span><span class="mf">0.</span><span class="p">,</span><span class="mi">0</span><span class="p">],[</span><span class="mf">3.</span><span class="p">,</span><span class="mf">3.</span><span class="p">]),</span> <span class="mf">0.3</span><span class="p">)</span>
<span class="n">sgxb</span><span class="p">,</span> <span class="n">sgyb</span> <span class="o">=</span> <span class="n">make_sample</span><span class="p">(</span><span class="mi">1000</span><span class="p">,</span> <span class="p">([</span><span class="mf">1.</span><span class="p">,</span><span class="mi">1</span><span class="p">],[</span><span class="mf">4.</span><span class="p">,</span><span class="mf">4.</span><span class="p">]),</span> <span class="mf">0.3</span><span class="p">)</span>
<span class="n">sgxc</span><span class="p">,</span> <span class="n">sgyc</span> <span class="o">=</span> <span class="n">make_sample</span><span class="p">(</span><span class="mi">1000</span><span class="p">,</span> <span class="p">([</span><span class="mf">5.</span><span class="p">,</span><span class="mf">5.</span><span class="p">],[</span><span class="o">-</span><span class="mf">2.</span><span class="p">,</span><span class="o">-</span><span class="mf">2.</span><span class="p">]),</span> <span class="mf">0.6</span><span class="p">)</span>
<span class="n">sgxd</span><span class="p">,</span> <span class="n">sgyd</span> <span class="o">=</span> <span class="n">make_sample</span><span class="p">(</span><span class="mi">1000</span><span class="p">,</span> <span class="p">([</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span><span class="mf">3.</span><span class="p">],[</span><span class="mf">3.</span><span class="p">,</span><span class="o">-</span><span class="mf">1.</span><span class="p">]),</span> <span class="mf">0.3</span><span class="p">)</span>
<span class="n">sgx</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">concatenate</span><span class="p">([</span><span class="n">sgxa</span><span class="p">,</span><span class="n">sgxb</span><span class="p">,</span><span class="n">sgxc</span><span class="p">,</span><span class="n">sgxd</span><span class="p">])</span>
<span class="n">sgy</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">concatenate</span><span class="p">([</span><span class="n">sgya</span><span class="p">,</span><span class="n">sgyb</span><span class="p">,</span><span class="n">sgyc</span><span class="p">,</span><span class="n">sgyd</span><span class="p">])</span>
</pre>
</div>
</div>
</div>
</div>
</div>
<div class="cell border-box-sizing code_cell rendered">
<div class="input">
<div class="prompt input_prompt">
In [12]:
</div>
<div class="inner_cell">
<div class="input_area">
<div class=" highlight hl-ipython3">
<pre><span></span><span class="n">plt</span><span class="o">.</span><span class="n">scatter</span><span class="p">(</span><span class="n">sgx</span><span class="p">[:,</span><span class="mi">0</span><span class="p">],</span> <span class="n">sgx</span><span class="p">[:,</span><span class="mi">1</span><span class="p">],</span> <span class="n">alpha</span><span class="o">=</span><span class="mf">0.5</span><span class="p">,</span> <span class="n">c</span><span class="o">=</span><span class="n">sgy</span><span class="p">)</span>
<span class="n">plt</span><span class="o">.</span><span class="n">xlabel</span><span class="p">(</span><span class="s1">'x1'</span><span class="p">)</span>
<span class="n">plt</span><span class="o">.</span><span class="n">ylabel</span><span class="p">(</span><span class="s1">'x2'</span><span class="p">)</span>
</pre>
</div>
</div>
</div>
</div>
<div class="output_wrapper">
<div class="output">
<div class="output_area">
<div class="prompt output_prompt">
Out[12]:
</div>
<div class="output_text output_subarea output_execute_result">
<pre>Text(0, 0.5, 'x2')</pre>
</div>
</div>
<div class="output_area">
<div class="prompt">
</div>
<div class="output_png output_subarea ">
<img src="data:image/png;base64,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
"/>
</div>
</div>
</div>
</div>
</div>
<div class="cell border-box-sizing text_cell rendered">
<div class="prompt input_prompt">
</div>
<div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<p>
Now, let's build a small network and see if we can classify that.
</p>
</div>
</div>
</div>
<div class="cell border-box-sizing code_cell rendered">
<div class="input">
<div class="prompt input_prompt">
In [13]:
</div>
<div class="inner_cell">
<div class="input_area">
<div class=" highlight hl-ipython3">
<pre><span></span><span class="n">mlp</span> <span class="o">=</span> <span class="n">MLPClassifier</span><span class="p">(</span><span class="n">hidden_layer_sizes</span><span class="o">=</span><span class="p">(</span><span class="mi">3</span><span class="p">,),</span> <span class="n">activation</span><span class="o">=</span><span class="s1">'relu'</span><span class="p">,</span> <span class="n">max_iter</span><span class="o">=</span><span class="mi">10000</span><span class="p">,</span> <span class="n">random_state</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
<span class="n">mlp</span><span class="o">.</span><span class="n">fit</span><span class="p">(</span><span class="n">sgx</span><span class="p">,</span><span class="n">sgy</span><span class="p">)</span>
</pre>
</div>
</div>
</div>
</div>
<div class="output_wrapper">
<div class="output">
<div class="output_area">
<div class="prompt output_prompt">
Out[13]:
</div>
<div class="output_text output_subarea output_execute_result">
<pre>MLPClassifier(activation='relu', alpha=0.0001, batch_size='auto', beta_1=0.9,
beta_2=0.999, early_stopping=False, epsilon=1e-08,
hidden_layer_sizes=(3,), learning_rate='constant',
learning_rate_init=0.001, max_iter=10000, momentum=0.9,
n_iter_no_change=10, nesterovs_momentum=True, power_t=0.5,
random_state=1, shuffle=True, solver='adam', tol=0.0001,
validation_fraction=0.1, verbose=False, warm_start=False)</pre>
</div>
</div>
</div>
</div>
</div>
<div class="cell border-box-sizing code_cell rendered">
<div class="input">
<div class="prompt input_prompt">
In [14]:
</div>
<div class="inner_cell">
<div class="input_area">
<div class=" highlight hl-ipython3">
<pre><span></span><span class="n">plot_result</span><span class="p">(</span><span class="n">sgx</span><span class="p">,</span><span class="n">sgy</span><span class="p">,</span><span class="n">linrange</span><span class="o">=</span><span class="p">(</span><span class="o">-</span><span class="mi">4</span><span class="p">,</span><span class="mi">7</span><span class="p">,</span><span class="mi">201</span><span class="p">))</span>
</pre>
</div>
</div>
</div>
</div>
<div class="output_wrapper">
<div class="output">
<div class="output_area">
<div class="prompt">
</div>
<div class="output_png output_subarea ">
<img src="data:image/png;base64,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
"/>
</div>
</div>
</div>
</div>
</div>
<div class="cell border-box-sizing text_cell rendered">
<div class="prompt input_prompt">
</div>
<div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<p>
The network does not have enough parameters to fit the training data.
</p>
<p>
<strong>
This is underfitting.
</strong>
</p>
<p>
Still, it does quite a good job with its three neurons.
</p>
<p>
Let's increase the number of neurons on the hidden layer a bit:
</p>
</div>
</div>
</div>
<div class="cell border-box-sizing code_cell rendered">
<div class="input">
<div class="prompt input_prompt">
In [15]:
</div>
<div class="inner_cell">
<div class="input_area">
<div class=" highlight hl-ipython3">
<pre><span></span><span class="n">mlp</span> <span class="o">=</span> <span class="n">MLPClassifier</span><span class="p">(</span><span class="n">hidden_layer_sizes</span><span class="o">=</span><span class="p">(</span><span class="mi">5</span><span class="p">,),</span> <span class="n">activation</span><span class="o">=</span><span class="s1">'relu'</span><span class="p">,</span> <span class="n">max_iter</span><span class="o">=</span><span class="mi">10000</span><span class="p">,</span> <span class="n">random_state</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
<span class="n">mlp</span><span class="o">.</span><span class="n">fit</span><span class="p">(</span><span class="n">sgx</span><span class="p">,</span><span class="n">sgy</span><span class="p">)</span>
<span class="n">plot_result</span><span class="p">(</span><span class="n">sgx</span><span class="p">,</span><span class="n">sgy</span><span class="p">,</span><span class="n">linrange</span><span class="o">=</span><span class="p">(</span><span class="o">-</span><span class="mi">4</span><span class="p">,</span><span class="mi">7</span><span class="p">,</span><span class="mi">201</span><span class="p">))</span>
</pre>
</div>
</div>
</div>
</div>
<div class="output_wrapper">
<div class="output">
<div class="output_area">
<div class="prompt">
</div>
<div class="output_png output_subarea ">
<img src="data:image/png;base64,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
"/>
</div>
</div>
</div>
</div>
</div>
<div class="cell border-box-sizing text_cell rendered">
<div class="prompt input_prompt">
</div>
<div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<p>
Wow! 5 neurons is already enough to fit the data, but we're a bit lucky. With a different topology we could have missed a patch.
</p>
<p>
Let's increase the complexity of the model even further:
</p>
</div>
</div>
</div>
<div class="cell border-box-sizing code_cell rendered">
<div class="input">
<div class="prompt input_prompt">
In [16]:
</div>
<div class="inner_cell">
<div class="input_area">
<div class=" highlight hl-ipython3">
<pre><span></span><span class="n">mlp</span> <span class="o">=</span> <span class="n">MLPClassifier</span><span class="p">(</span><span class="n">hidden_layer_sizes</span><span class="o">=</span><span class="p">(</span><span class="mi">50</span><span class="p">,</span><span class="mi">50</span><span class="p">,</span><span class="mi">50</span><span class="p">),</span> <span class="n">activation</span><span class="o">=</span><span class="s1">'relu'</span><span class="p">,</span> <span class="n">max_iter</span><span class="o">=</span><span class="mi">10000</span><span class="p">,</span> <span class="n">random_state</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
<span class="n">mlp</span><span class="o">.</span><span class="n">fit</span><span class="p">(</span><span class="n">sgx</span><span class="p">,</span><span class="n">sgy</span><span class="p">)</span>
<span class="n">plot_result</span><span class="p">(</span><span class="n">sgx</span><span class="p">,</span><span class="n">sgy</span><span class="p">,</span><span class="n">linrange</span><span class="o">=</span><span class="p">(</span><span class="o">-</span><span class="mi">4</span><span class="p">,</span><span class="mi">7</span><span class="p">,</span><span class="mi">201</span><span class="p">))</span>
</pre>
</div>
</div>
</div>
</div>
<div class="output_wrapper">
<div class="output">
<div class="output_area">
<div class="prompt">
</div>
<div class="output_png output_subarea ">
<img src="data:image/png;base64,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
"/>
</div>
</div>
</div>
</div>
</div>
<div class="cell border-box-sizing text_cell rendered">
<div class="prompt input_prompt">
</div>
<div class="inner_cell">
<div class="text_cell_render border-box-sizing rendered_html">
<p>
Still no overfitting. The network now shows a smooth boundary, and I'm pretty sure it would be able to adapt further if needed.
</p>
<p>
Now
<a href="https://thedatafrog.com/overfitting-illustrated#conclusion">
let's go back and wrap up!
</a>
</p>
</div>
</div>
</div>
</div>
<!-- mathjax -->
<script async="" id="MathJax-script" src="https://cdn.jsdelivr.net/npm/mathjax@3/es5/tex-mml-chtml.js">
</script>
<script src="https://polyfill.io/v3/polyfill.min.js?features=es6">
</script>
<script>
MathJax = {
tex: {
inlineMath: [['$', '$'], ['\\(', '\\)']]
},
svg: {
fontCache: 'global'
}
};
</script>
</body>
</html>