Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
Machine Learning with Clojure and Apache Spark
Search
Eric Weinstein
October 25, 2016
Technology
1
330
Machine Learning with Clojure and Apache Spark
Slides for my EuroClojure 2016 talk on machine learning.
Eric Weinstein
October 25, 2016
Tweet
Share
More Decks by Eric Weinstein
See All by Eric Weinstein
Interview Them Where They Are
ericqweinstein
0
98
Value Your Types!
ericqweinstein
0
57
Being Good: An Introduction to Robo- and Machine Ethics
ericqweinstein
1
1.6k
What If...?: Ruby 3
ericqweinstein
1
180
Infinite State Machine
ericqweinstein
1
90
Do Androids Dream of Electronic Dance Music?
ericqweinstein
1
73
Machine Learning with Elixir and Phoenix
ericqweinstein
1
840
Domo Arigato, Mr. Roboto: Machine Learning with Ruby
ericqweinstein
1
1.2k
A Nil Device, A Lonely Operator, and a Voyage to the Void Star
ericqweinstein
1
780
Other Decks in Technology
See All in Technology
開発パフォーマンスを最大化するための開発体制
ham0215
2
410
私が trocco を推す理由
__allllllllez__
1
220
元インフラエンジニアに成る / Human Resources to Human Relations
bobtani
4
920
On Your Data を超えていく!
hirotomotaguchi
2
680
チームでロジカルシンキングに改めて向き合っている話 〜学習環境と実践⽅法〜
sansantech
PRO
3
2.5k
ServiceNow Knowledge 24の歩き方 EYストラテジー・アンド・コンサルティング
manarobot
0
200
反実仮想機械学習とは何か
usaito
PRO
11
4.6k
エンジニア候補者向け資料2024.04.24.pdf
macloud
0
3.3k
require(ESM)とECMAScript仕様
uhyo
3
660
Além do else! Categorizando Pokemóns com Pattern Matching no JavaScript
wmsbill
0
620
自己改善からチームを動かす! 「セルフエンジニアリングマネージャー」のすゝめ
shoota
6
670
Postman v10リリース後を振り返る / Looking back at Postman v10 after release
yokawasa
1
160
Featured
See All Featured
Designing on Purpose - Digital PM Summit 2013
jponch
110
6.5k
The Cult of Friendly URLs
andyhume
74
5.7k
Visualizing Your Data: Incorporating Mongo into Loggly Infrastructure
mongodb
34
8.9k
VelocityConf: Rendering Performance Case Studies
addyosmani
320
23k
Let's Do A Bunch of Simple Stuff to Make Websites Faster
chriscoyier
501
140k
No one is an island. Learnings from fostering a developers community.
thoeni
16
2.1k
jQuery: Nuts, Bolts and Bling
dougneiner
59
7.1k
What's new in Ruby 2.0
geeforr
337
31k
Designing for Performance
lara
601
67k
Faster Mobile Websites
deanohume
299
30k
It's Worth the Effort
3n
180
27k
A Tale of Four Properties
chriscoyier
151
22k
Transcript
Machine Learning with Clojure and Apache Spark ;; Eric Weinstein
;; EuroClojure 2016 ;; Bratislava, Slovakia ;; 25 October 2016
for Joshua
Part 0: Hello!
About Me (def eric-weinstein {:employer "Hulu" :github "ericqweinstein" :twitter "ericqweinstein"
:website "ericweinste.in"}) 30% off with EURORUBY30!
Agenda • Machine learning • Apache Spark • Flambo vs.
Sparkling • DL4J, deep learning, and convolutional neural networks
Part 1: ⚡✨
What’s machine learning?
In a word:
Generalization
What’s Supervised Learning? Classification or regression, generalizing from labeled data
to unlabeled data
What’s Apache Spark? Apache Spark is an open-source cluster computing
framework; its parallelism makes it ideal for processing large data sets, and in ML, the more data, the better!
Some Spark Terminology • RDD: Resilient Distributed Dataset • Dataset:
RDD + Spark SQL execution engine • DataFrame: Dataset organized into named columns
Our Data • Police stop data for the city of
Los Angeles, California in 2015 • 4 features, ~600,000 instances • http://bit.ly/2f9jVwn
Features && Labels • Sex (Male | Female) • Race
(American Indian | Asian | Black | Hispanic | White | Other) • Stop type (Pedestrian | Vehicle) • Post-stop activity (Yes | No)
Features && Labels • Sex (Male | Female) • Race
(American Indian | Asian | Black | Hispanic | White | Other) • Stop type (Pedestrian | Vehicle) • Post-stop activity (Yes | No)
Decision Trees X[0] <= 0.5 gini = 0.4033 samples =
139572 value = [100477, 39095] X[1] <= 5.5 gini = 0.4318 samples = 102419 value = [70118, 32301] True X[1] <= 5.5 gini = 0.2989 samples = 37153 value = [30359, 6794] False X[1] <= 4.5 gini = 0.4399 samples = 96665 value = [65083, 31582] gini = 0.2187 samples = 5754 value = [5035, 719] X[1] <= 3.5 gini = 0.4483 samples = 78400 value = [51805, 26595] gini = 0.397 samples = 18265 value = [13278, 4987] X[1] <= 2.5 gini = 0.4324 samples = 51662 value = [35328, 16334] gini = 0.473 samples = 26738 value = [16477, 10261] X[1] <= 0.5 gini = 0.4406 samples = 48927 value = [32894, 16033] gini = 0.1959 samples = 2735 value = [2434, 301] gini = 0.4658 samples = 65 value = [41, 24] gini = 0.4406 samples = 48862 value = [32853, 16009] X[1] <= 3.5 gini = 0.3067 samples = 34817 value = [28234, 6583] gini = 0.1643 samples = 2336 value = [2125, 211] X[1] <= 2.5 gini = 0.2796 samples = 15786 value = [13133, 2653] X[1] <= 4.5 gini = 0.3277 samples = 19031 value = [15101, 3930] X[1] <= 0.5 gini = 0.2921 samples = 13985 value = [11501, 2484] gini = 0.1701 samples = 1801 value = [1632, 169] gini = 0.426 samples = 26 value = [18, 8] gini = 0.2918 samples = 13959 value = [11483, 2476] gini = 0.3747 samples = 9522 value = [7144, 2378] gini = 0.2732 samples = 9509 value = [7957, 1552]
Part 2: A Tale of Two DSLs vs. ✨✨ Image
credit: Adventure Time
Flambo Example (defn make-spark-context "Creates the Apache Spark context using
the Flambo DSL." [] (-> (conf/spark-conf) (conf/master "local") (conf/app-name "euroclojure") (f/spark-context)))
Sparkling Example (defn make-spark-context "Creates the Apache Spark context using
the Sparkling DSL." [] (-> (conf/spark-conf) (conf/master "local") (conf/app-name "euroclojure") (spark/spark-context)))
Straight Spark (def model (DecisionTree/trainClassifier training 2 categorical-features- info "gini"
5 32)) ; max depth: 5, max leaves: 32 (defn predict [p] ; LabeledPoint (let [prediction (.predict model (.features p))] [(.label p) prediction]))
Accuracy: 0.77352
Part 3: Deep Learning
What’s Deep Learning? • Neural networks (computational architecture modeled after
the human brain) • Neural networks with many layers (> 1 hidden layer, but in practice, can be hundreds) • The vanishing/exploding gradient problem
Vanishing && Gradients
Image credit for all ConvNet images: https://deeplearning4j.org/convolutionalnets
Max Pooling/Downsampling
Alternating Layers
Our Data Image credit: http://digitalmedia.fws.gov/cdm/
What’s DL4J? • DL4J == Deep Learning 4 Java, a
library (for Java, unsurprisingly) • Examples on GitHub: https://github.com/ deeplearning4j/deeplearning4j • ConvNet worked example: http://bit.ly/2eBM8ss
DL4J Example (def nn-conf (-> (NeuralNetConfiguration$Builder.) ;; Some values omitted
for space (.activation "relu") (.learningRate 0.0001) (.weightInit (WeightInit/XAVIER)) (.optimizationAlgo OptimizationAlgorithm/STOCHASTIC_GRADIENT_DESCENT) (.updater Updater/RMSPROP) (.momentum 0.9) (.list) (.layer 0 conv-init) (.layer 1 (max-pool "maxpool1" (int-array [2 2]))) (.layer 2 (conv-5x5 "cnn2" 100 (int-array [5 5]) (int-array [1 1]) 0)) (.layer 3 (max-pool "maxpool2" (int-array [2 2]))) (.layer 4 (fully-connected 500)) (.layer 5 output-layer) (.build)))
How’d We Do? • Accuracy: 0.375 • Precision: 0.3333 •
Recall: 0.375 • F1 Score: 0.3529
Summary • Clojure + Spark = • Flambo and Sparkling
are roughly equally powerful • Deep learning is super doable with Clojure (though Java interop is kind of a pain)
Takeaways (TL;DPA) • Contribute to Flambo and/or Sparkling! • Let’s
build or contribute to a nicer DSL for DL4J • https://github.com/ericqweinstein/euroclojure
None