forked from DJCordhose/ml-workshop
-
Notifications
You must be signed in to change notification settings - Fork 0
/
2019-intro.html
901 lines (718 loc) · 27.2 KB
/
2019-intro.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
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Intro</title>
<meta name="description" content="A framework for easily creating beautiful presentations using HTML">
<meta name="author" content="Hakim El Hattab">
<meta name="apple-mobile-web-app-capable" content="yes"/>
<meta name="apple-mobile-web-app-status-bar-style" content="black-translucent"/>
<meta name="viewport"
content="width=device-width, initial-scale=1.0, maximum-scale=1.0, user-scalable=no, minimal-ui">
<link rel="stylesheet" href="reveal.js/css/reveal.css">
<!--<link rel="stylesheet" href="reveal.js/css/theme/white.css" id="theme">-->
<!--<link rel="stylesheet" href="reveal.js/css/theme/black.css" id="theme">-->
<!--<link rel="stylesheet" href="reveal.js/css/theme/night.css" id="theme">-->
<!--<link rel="stylesheet" href="reveal.js/css/theme/simple.css" id="theme">-->
<link rel="stylesheet" href="reveal.js/css/theme/solarized.css" id="theme">
<!-- Code syntax highlighting -->
<link rel="stylesheet" href="reveal.js/lib/css/zenburn.css">
<style>
/*pre code {*/
/*display: block;*/
/*padding: 0.5em;*/
/*background: #FFFFFF !important;*/
/*color: #000000 !important;*/
/*}*/
.right-img {
margin-left: 10px !important;
float: right;
height: 500px;
}
.todo:before {
content: 'TODO: ';
}
.todo {
color: red !important;
}
code span.line-number {
color: lightcoral;
}
.reveal pre code {
max-height: 1000px !important;
}
img {
border: 0 !important;
box-shadow:0 0 0 0 !important;
}
.reveal {
-ms-touch-action: auto !important;
touch-action: auto !important;
}
.reveal h2,
.reveal h3,
.reveal h4 {
letter-spacing: 2px;
font-family: 'Calibri', sans-serif;
/* font-family: 'Times New Roman', Times, serif; */
font-weight: bold;
color: black;
font-style: italic;
letter-spacing: -2px;
text-transform: none !important;
}
.reveal em {
font-weight: bold;
}
.reveal .step-subtitle h1 {
letter-spacing: 1px;
}
.reveal .step-subtitle h2,
.reveal .step-subtitle h3 {
text-transform: none;
font-style: italic;
font-weight: normal;
/* font-weight: 400; */
/* font-family: 'Amiri', serif; */
font-family: 'Lobster', serif;
letter-spacing: 1px;
color: #2aa198;
text-decoration: underline;
}
.reveal .front-page h1,
.reveal .front-page h2 {
font-family: "League Gothic";
font-style: normal;
text-transform: uppercase !important;
letter-spacing: 1px;
}
.reveal .front-page h1 {
font-size: 2.5em !important;
}
.reveal .highlight {
background-color: #D3337B;
color: white;
}
.reveal section img {
background: none;
}
.reveal img.with-border {
border: 1px solid #586e75 !important;
box-shadow: 3px 3px 1px rgba(0, 0, 0, 0.15) !important;
}
.reveal li {
margin-bottom: 8px;
}
/* For li's that use FontAwesome icons as bullet-point */
.reveal ul.fa-ul li {
list-style-type: none;
}
.reveal {
color: black !important;
}
</style>
<!-- Printing and PDF exports -->
<script>
var link = document.createElement('link');
link.rel = 'stylesheet';
link.type = 'text/css';
var printMode = window.location.search.match(/print-pdf/gi);
link.href = printMode ? 'reveal.js/css/print/pdf.css' : 'reveal.js/css/print/paper.css';
document.getElementsByTagName('head')[0].appendChild(link);
</script>
<!--[if lt IE 9]>
<script src="reveal.js/lib/js/html5shiv.js"></script>
<![endif]-->
</head>
<body>
<div class="reveal">
<div class="slides">
<!-- JFokus Talk
3 hours
https://www.jfokus.se/jfokus19/talks/2634
Hands-on-lab: Grundlagen von Machine Learning
In dieser Hands-On-Session bekommst du eine Einführung in die Grundlagen von Machine Learning. Wir machen dabei Experimente auf
Papier und erste Erfahrungen mit der Python-Bibliothek Scikit-learn. Du brauchst aber weder Erfahrung in Machine Learning, noch
in Python oder Scikit-learn. Du brauchst lediglich ein Laptop auf dem der Chrome Browser installiert ist.
-->
<section data-markdown class="variations">
<textarea data-template>
* NVE Workshop
* ML Summit
</textarea>
</section>
<section data-markdown class="preparation">
<textarea data-template>
### Preparation Workshop
* exercise/2019-workshop-exercise-supervised.docx ausdrucken für alle Teilnehmer
</textarea>
</section>
<!-- <section data-markdown style="font-size: xx-large">
<textarea data-template>
## Hands-on-lab: Machine Learning for Software Developers
If you are bored
1. #wifi Name: XXX, pwd: XXX
1. Open this slide deck: http://bit.ly/ml-jfokus-hands-on
1. Make sure you are ready to work with Colab
* open https://colab.research.google.com/github/djcordhose/ai/blob/master/notebooks/sklearn/dt-intro.ipynb in Chrome
* make it run using the "Run All" command from the "Runtime" menu
* save the notebook in your Drive
</textarea>
</section> -->
<!-- <section>
<br>
<br>
<h2>Grundlagen von Machine Learning</h2>
<br>
<br>
<p><a target="_blank" href="xxx">
ML Summit, Berlin, October 2019
</a></p>
<h4><a href="http://zeigermann.eu">Oliver Zeigermann</a> /
<a href="http://twitter.com/djcordhose">@DJCordhose</a>
</h4>
<p><small><a href="http://bit.ly/xxx">
http://bit.ly/xxxx
</a></small></p>
</section> -->
<section data-markdown class="todo">
<textarea data-template>
Intro Wie bei jfokus: 2019_jfokus_workshop.html ohne TensorFlow 0,5
</textarea>
</section>
<section data-markdown class="todo">
<textarea data-template>
Depending on what your background is, machine learning can have different challenges. If you come from data science the
idea of treating your working like a software project that needs to go through different phases - one of them being
production - and cycles might be new to you. A notebook is great to try things out and produce results from data once,
but might not be the right tool when your project lives for a longer amount of time. If you, however, come from software
development, understanding that data is the critical factor and if it is lacking in either quantity or quality you might
not be able to have a working model at all. This makes the process for creating a good machine learning model pretty
different from standard processes in software development.
</textarea>
</section>
<section data-markdown class="todo">
<textarea data-template>
### Abschluss
<img src='img/ML-strategy-helper.png'>
</textarea>
</section>
<!-- <section data-markdown class="todo">
<textarea data-template>
</textarea>
</section> -->
<section data-markdown>
<textarea data-template>
### Our plan for today
1. Nature of Machine Learning
1. Data Preparation (Optional)
1. Implementation using Scikit-learn
1. Making sure we really generalize to unknown data
</textarea>
</section>
<section data-markdown class="local no-fragment"">
<textarea data-template>
## Stellt euch euren Nachbarn vor
* Was macht ihr?
* Was wollt ihr mit Machine Learning erreichen?
* Was wisst ihr schon?
* Falls noch nicht passiert, helft euren Nachbarn, diese Folien zu bekommen http://bit.ly/xxx und das Notebook https://colab.research.google.com/github/djcordhose/ai/blob/master/notebooks/sklearn/dt-intro.ipynb zum laufen zu bringen (siehe erste Folie)
</textarea>
</section>
<section data-markdown>
<textarea data-template>
### Our Scenario for today
* We are CTO of a highly innovative Car Insurance Company
* Different from other insurance companies we determine the rate by the actual number of accidents per customer
* _Objective: how many accidents will prospective customers have?_
<img src='img/pixabay/accident-151668_1280.png' height="300px">
</textarea>
</section>
<section data-markdown>
<textarea data-template>
### Three Inputs for Prediction
<a href='html/calculator.html'>
<img src='img/insurance/calculator.png' height="450">
</a>
<p><small>
<a href='html/calculator.html' target="_blank">
https://djcordhose.github.io/ai/html/calculator.html</a></small>
</small></p>
</textarea>
</section>
<section data-markdown>
<textarea data-template>
## How to solve this programmatically?
</textarea>
</section>
<section data-markdown>
<textarea data-template>
<img src='img/classic-development.jpg'>
</textarea>
</section>
<section data-markdown>
<textarea data-template>
### Exploration
<img src='img/insurance/sns-pairplot.png' height="550px">
</textarea>
</section>
<section>
<h3>Programmer's approach: Code Rules by Hand</h3>
<div class="fragment">
<pre><code contenteditable data-trim class="line-numbers python">
if age < 25:
if speed > 140:
return red # young people, fast cars: high risk
else:
return yellow # young people: medium risk
</code></pre>
</div>
<div class="fragment">
<pre><code contenteditable data-trim class="line-numbers python">
if age > 75:
return red # old people: high risk
</code></pre>
</div>
<div class="fragment">
<pre><code contenteditable data-trim class="line-numbers python">
if miles_per_year > 30:
return red # a lot of driving: high risk
if miles_per_year > 20:
return yellow # a bit of driving: medium risk
</code></pre>
</div>
<div class="fragment">
<pre><code contenteditable data-trim class="line-numbers python">
return green # otherwise: low risk
</code></pre>
</div>
<p><small>
<a href='https://colab.research.google.com/github/djcordhose/ai/blob/master/notebooks/sklearn/classic-code.ipynb>'>https://colab.research.google.com/github/djcordhose/ai/blob/master/notebooks/sklearn/classic-code.ipynb</a>
</small></p>
</section>
<section data-markdown>
<textarea data-template>
### How good is this?
<!-- <p class="fragment"><em>How well have we mastered the domain?</em></p> -->
* Is it better than guessing?
* Are all the rules correct?
* Are some missing?
* How would we even know?
</textarea>
</section>
<section>
<h3>Different Approach</h3>
<div style="max-width: 50%; float: left;">
<img src='img/insurance/df_head.jpg' height="450">
</div>
<div style="max-width: 50%; float: right;">
<br>
<br>
<br>
<ul>
<li>0 - red: many accidents</li>
<li>1 - green: few or no accidents</li>
<li>2 - yellow: in the middle</li>
</ul>
</div>
<p style="clear: both;">
<em>
Looking at the problem from the perspective of our data
</em>
</p>
</section>
<section>
<h3>Let's check: Is the data we have any good?</h3>
<img src="img/insurance/all.png" height="500px" class="fragment">
<small class="fragment">Would we able to rank me (47) for a car having 100 mph top speed, driving 10k miles per year?</small>
</section>
<section data-markdown class="workshop">
<textarea data-template>
### Would we be able to separate different classes by drawing a few lines?
</textarea>
</section>
<section data-markdown class="workshop exercise">
<textarea data-template>
### Paper Exercise #1 - Understanding the Supervised Learning Approach
_team up and discuss with your neighbors_
* Manually cluster areas of different customer types on the <em>upper part</em> of the handout
* Be sure to use a ballpen and apply pressure when drawing
<a href='/exercise/2019-workshop-exercise.pdf'>Exercise as PDF</a>
</textarea>
</section>
<section data-markdown class="workshop">
<textarea data-template>
### Two Sample Solutions
<img src='img/decision-boundaries/decision-boundaries-train.jpg'>
<small>
Are they the same? What is the key difference?
</small>
</textarea>
</section>
<!-- <section class="workshop">
<h3>Manual Classification - red</h3>
<img src="img/decision-boundaries/decision-boundaries-1.jpg">
</section>
<section class="workshop">
<h3>Manual Classification - yellow</h3>
<img src="img/decision-boundaries/decision-boundaries-2.jpg">
</section>
<section class="workshop">
<h3>Manual Classification - green</h3>
<img src="img/decision-boundaries/decision-boundaries-3.jpg">
</section> -->
<section class="workshop">
<h3>Decision Bounaries</h3>
<img src="img/decision-boundaries/decision-boundaries-4.jpg">
</section>
<section data-markdown class="workshop">
<textarea data-template>
## Core Question
### Can we automate this process of drawing Decision Bondaries?
</textarea>
</section>
<section data-markdown>
<textarea data-template>
<img src='img/supervised-ml.jpg' height="650px">
</textarea>
</section>
<section data-markdown class="workshop">
<textarea data-template>
### Step I
## Data Preparation
</textarea>
</section>
<section data-markdown class="workshop">
<textarea data-template>
### Data is King
_collecting data might be the hardest part of the job_
* but also the most important
* no data, not good quality or quantity => no supervised machine learning
* if we have a simulator, reinforcement learning might be an option
</textarea>
</section>
<section data-markdown class="workshop">
<textarea data-template>
### Clean Data and Select Features
<img src='img/insurance/data.png' height="500">
</textarea>
</section>
<section data-markdown class="workshop exercise">
<textarea data-template>
### Exercise on Paper #2 - Data Cleaning and Feature Selection
_Data Cleaning_
* What errors do you find in the data? Mark on paper and describe
* How to deal with those errors?
_Feature Selection_
* Which column would you predict?
* Which columns would you use as input for training?
* Would you use all columns? If not, why?
<!--
- Datendopplung Zeile 4/5 => eine löschen
- Fehlender Wert Zeile 16 => Zeile löschen oder Durchschnitt
- Spalte 'state' mit mehr als 50% fehlenden Werten
- Zeile 23 Ausreißer: Zeile löschen oder auf plausiblen Wert korrigieren
- Zeile 24: Califorina
-->
</textarea>
</section>
<section data-markdown style="font-size: xx-large" class="workshop">
<textarea data-template>
### Results: Data Cleaning und Feature Selection
_Data Cleaning_
* Typos: Califorina
* Outliers: Delete line or replace with decent value
* Doubles: Delete
* Missing Value: Delete line or replace with imputed value
_Feature Selection_
* Make sure which value to predict
* Row missig more than 50% of values: do not use
* Explore dependencies to decide what to use for training input
</textarea>
</section>
<section data-markdown class="detail">
<textarea data-template>
### Final Check on Collected Data
<img src='img/insurance/df_describe.png' height="500">
</textarea>
</section>
<section data-markdown class="workshop">
<textarea data-template>
### Step II
## Training
</textarea>
</section>
<section data-markdown>
<textarea data-template>
### Revisiting our Rule based approach
Plotting the predictions as a background
<img src='img/decision-boundaries/5-prog.png' class="fragment" height="450px">
<br>
<small class="fragment">approx. 43% predictions correct</small>
</textarea>
</section>
<section data-markdown>
<textarea data-template>
## Do we really have to write those rules by hand?
</textarea>
</section>
<section data-markdown>
<textarea data-template>
### Decision Trees can learn such rules
<img src="img/dt/dt-all.png" height="500px">
</textarea>
</section>
<section>
<h3>Code in Scikit-learn</h3>
<p>Training</p>
<pre><code contenteditable data-trim class="fragment line-numbers python">
from sklearn.tree import DecisionTreeClassifier
clf = DecisionTreeClassifier()
clf.fit(X_train, y_train)
</code></pre>
<p>Prediction</p>
<pre><code contenteditable data-trim class="fragment line-numbers python">
y_pred = clf.predict(input)
</code></pre>
<small>
<a href='https://scikit-learn.org/stable/modules/classes.html'>https://scikit-learn.org/stable/modules/classes.html</a>
<!-- <br>
<a href='https://colab.research.google.com/github/djcordhose/ai/blob/master/notebooks/sklearn/overview.ipynb'>
https://colab.research.google.com/github/djcordhose/ai/blob/master/notebooks/sklearn/overview.ipynb</a> -->
</small>
</section>
<section data-markdown class="detail">
<textarea data-template>
### How do they learn?
<img src="img/dt/cart.png" height="500px">
<small>
http://scikit-learn.org/stable/modules/tree.html#tree-algorithms-id3-c4-5-c5-0-and-cart
</small>
</textarea>
</section>
<section>
<h3>Decision Trees</h3>
<div>
<img src="img/decision-boundaries/dt-sweet-test.png">
<small>Up to 70% accuracy on unknown data</small>
</div>
</section>
<!-- <section data-markdown>
<textarea data-template>
### Working with Colab Notebooks
https://colab.research.google.com
</textarea>
</section> -->
<section data-markdown>
<textarea data-template>
### Hands-On
_Run your first Colab Notbook_
* Open the notebook containing the training code
* Sign into your Google account or register a new one
* Execute cells until the notebook tells you to stop
* Train the model and see how it would classify yourself
* Save your notebook in your Google Drive (or on Github if you want to)
<small>
https://colab.research.google.com/github/djcordhose/ai/blob/master/notebooks/sklearn/dt-intro.ipynb
</small>
</textarea>
</section>
<section data-markdown class="workshop">
<textarea data-template>
### Step III
## Evaluation
</textarea>
</section>
<section data-markdown>
<textarea data-template>
### Machine Learning is all about Generalization
<img src='img/MELLI.jpg' height="500px">
</textarea>
</section>
<section data-markdown>
<textarea data-template>
### Split known data into training and test
<img class='fragment' src='img/insurance/generalization.jpg' height="550px">
</textarea>
</section>
<section data-markdown class="workshop exercise">
<textarea data-template>
### Exercise on Paper #3 - How well did you generalize?
* Turn the page and redraw the lines you printed through from the first page
* Get an estimate of the misclassified samples on both diagrams
* What do you think, is this result worse?
* What would you have done differently if you had known this is all about generalization?
</textarea>
</section>
<section id='overfitting'>
<h3>The Issue: Overfitting</h2>
<div>
<div style="float: left">
<img src="img/elements/80_percent.jpg" height="200" class="fragment" data-fragment-index='1'>
<p>
<small><em>Training Score</em></small>
</p>
</div>
<div style="float: left" class="fragment" data-fragment-index='5'>
<img src="img/elements/down.jpg" height="200">
</div>
<div style="float: left" class="fragment" data-fragment-index='4'>
<img src="img/elements/up.jpg" height="200">
</div>
<div style="float: left">
<img src="img/elements/70_percent.jpg" height="225" class="fragment" data-fragment-index='2'>
<p>
<small><em>Test Score</em></small>
</p>
</div>
</div>
<p style="clear: both" class="fragment" data-fragment-index='3'><em>Training and test scores clearly divert</em></p>
</section>
<section data-markdown>
<textarea data-template>
### Regularization
_Process to counter overfitting_
Each ML strategy has its own means of Regularization, e.g.
* KNN: more neighbors
* Decision Trees: reduce depth, use ensembles
* SVM: gamma (how many support vectors to use) and cost (of misclassified data)
* NN: Dropout, Batch Normalization, Reduced Capacity, Reduced Training Time
</textarea>
</section>
<section data-markdown>
<textarea data-template>
### Hands-On: Colab Notebook Part II
_Regularize your Decision tree_
* Continue with your notebook from where you stopped
* Execute the rest of the notebook
* Change the maximum depth of the decision tree to fight overfitting
* What is your best depth?
* How do the decision boundaries change?
* What are your best scores?
<small>
https://colab.research.google.com/github/djcordhose/ai/blob/master/notebooks/sklearn/dt-intro.ipynb
</small>
</textarea>
</section>
<section data-markdown class="workshop">
<textarea data-template>
### Remember the two sample solutions?
<img src='img/decision-boundaries/decision-boundaries-train.jpg'>
<small>
Are they the same? What is the key difference?
</small>
</textarea>
</section>
<section data-markdown class="workshop">
<textarea data-template>
### Now on test data
<img src='img/decision-boundaries/decision-boundaries-test.jpg'>
<small>
Which one is better? Why?
</small>
</textarea>
</section>
<section data-markdown class="workshop">
<textarea data-template>
### Where to go from here
* Find yourself some datasets to experiment with: https://toolbox.google.com/datasetsearch
* TensorFlow datasets (directly from Colab): https://github.com/tensorflow/datasets
* Watch my Deep Learning Crash Course: https://www.manning.com/livevideo/deep-learning-crash-course?a_aid=djcordhose&a_bid=e8e77cbf
</textarea>
</section>
<section style="font-size: xx-large">
<h2>Wrap-Up</h2>
<ul>
<li class="fragment">Machine Learning can learn from existing data and generalize from it
<li class="fragment">Sometimes, it can be an alternative to explicit, classic business logic
<li class="fragment">Classic Machine Learning strategies are often enough
<li class="fragment">Decision Trees and Random Forests are powerful strategies provided by Scikit-learn
<li class="fragment">Colab is a free service for executing your Python code using Scikit-learn and TensorFlow
</ul>
<br>
<p>
<em>JFokus 2019: Machine Learning for Software Developers</em>
<br>
<br>
<a href="http://zeigermann.eu">Oliver Zeigermann</a> / <a href="http://twitter.com/djcordhose">@DJCordhose</a></a>
<br>
<a href="http://bit.ly/ml-jfokus-hands-on">
http://bit.ly/ml-jfokus-hands-on
</a>
</p>
</section>
</div>
</div>
<script src="reveal.js/js/reveal.js"></script>
<script src="lib/jquery-2.2.4.js"></script>
<script>
$('section:not([data-background])').attr('data-background', "background/white.jpg");
</script>
<script>
const isLocal = window.location.hostname.indexOf('localhost') !== -1 ||
window.location.hostname.indexOf('127.0.0.1') !== -1;
if (isLocal && !printMode) {
} else {
// only applies to public version
$('.todo').remove();
$('.preparation').remove();
$('.local').remove();
}
Reveal.addEventListener( 'ready', function( event ) {
// applies to all versions
$('code').addClass('line-numbers');
$('.fragments li').addClass('fragment')
// make all links open in new tab
$('a').attr('target', '_blank')
if (isLocal && !printMode) {
// only applies to presentation version
Reveal.configure({ controls: false });
} else {
// only applies to public version
$('.fragment').removeClass('fragment');
}
} );
</script>
<script>
// Full list of configuration options available at:
// https://github.com/hakimel/reveal.js#configuration
Reveal.initialize({
controls: true,
progress: true,
history: true,
center: true,
width: 1100,
transition: 'fade', // none/fade/slide/convex/concave/zoom
// Optional reveal.js plugins
dependencies: [
{
src: 'reveal.js/lib/js/classList.js', condition: function () {
return !document.body.classList;
}
},
{
src: 'reveal.js/plugin/markdown/marked.js', condition: function () {
return !!document.querySelector('[data-markdown]');
}
},
{
src: 'reveal.js/plugin/markdown/markdown.js', condition: function () {
return !!document.querySelector('[data-markdown]');
}
},
{
src: 'reveal.js/plugin/highlight/highlight.js', async: true, condition: function () {
return !!document.querySelector('pre code');
}, callback: function () {
hljs.initHighlightingOnLoad();
}
},
{src: 'reveal.js/plugin/zoom-js/zoom.js', async: true},
{src: 'reveal.js/plugin/notes/notes.js', async: true},
// https://github.com/mikemiles86/reveal-line-numbers
{src: 'lib/js/line-numbers.js'}
]
});
</script>
</body>
</html>