Apache PredictionIO - Machine Learning 15minutes! #12

2017/05/27に開催された第12回 Machine Learning 15minutes! でのLT資料です。
「Apache PredictionIOのコミッタが語る Spark&Elasticsearch 機械学習基盤 」

Transcript

1. Open Source Machine Learning Server 15 Machine Learning minutes!
2. Apache PredictionIOͷίϛολ͕ޠΔ Spark&Elasticsearch ػցֶशج൫
3. ػցֶशͷ͓೰Έ • ֶशɾϞσϧσʔλͷετϨʔδ • ػցֶशͷ෼ࢄॲཧϑϨʔϜϫʔΫ • σʔλ༧ଌͷWebαʔϏεʢAPIʣԽ ղܾͷબ୒ࢶ → ఻͍͑ͨ͜ͱ
4. ػցֶशͷ͓೰Έ • ֶशɾϞσϧσʔλͷετϨʔδ • ػցֶशͷ෼ࢄॲཧϑϨʔϜϫʔΫ • σʔλ༧ଌͷWebαʔϏεʢAPIʣԽ ղܾͷબ୒ࢶ → ఻͍͑ͨ͜ͱ
5. Photo by Bernard Spragg. NZ About me
6. Profile Takahiro Hagino Bizreach גࣜձࣾϏζϦʔν • ٻਓݕࡧΤϯδϯʮελϯόΠʯ • AIࣨ
7. Open Source Machine Learning Server
8. ϏζϦʔνͰ͸ػցֶशΞϓϦέʔγϣϯͷ
   ։ൃɾӡ༻ج൫ͱͯ͠
   "QBDIF
   1SFEJDUJPO*0ʹऔΓ૊ΜͰ͍·͢ɻ
   ೥݄
   ฐ͔ࣾΒ໊͕ίϛολͱͯ͠ 1SFEJDUJPO*0ͷ։ൃʹࢀՃ͢Δ͜ͱʹͳΓ·ͨ͠ɻ Shinsuke Sugaya Naoki
 Takezoe Takako
 Shimamoto Takahiro Hagino
9. ϏζϦʔνͰ͸ػցֶशΞϓϦέʔγϣϯͷ
   ։ൃɾӡ༻ج൫ͱͯ͠
   "QBDIF
   1SFEJDUJPO*0ʹऔΓ૊ΜͰ͍·͢ɻ
   ೥݄
   ฐ͔ࣾΒ໊͕ίϛολͱͯ͠ 1SFEJDUJPO*0ͷ։ൃʹࢀՃ͢Δ͜ͱʹͳΓ·ͨ͠ɻ Shinsuke Sugaya Naoki
 Takezoe Takako
 Shimamoto Takahiro Hagino
11. x ML ٻ৬ऀͱٻਓͷϚονϯά Search Quality and Recommendation ೥ऩਪఆ Salary Prediction ۀछɾ৬छਪఆ Job Category Prediction ٻਓಛ௃ਪఆ Prediction of Job Characteristics
12. ٻ৬ऀͱٻਓͷϚονϯά Search Quality and Recommendation ೥ऩਪఆ Salary Prediction ۀछɾ৬छਪఆ Job Category Prediction ٻਓಛ௃ਪఆ Prediction of Job Characteristics x ML
13. ٻ৬ऀͱٻਓͷϚονϯά Search Quality and Recommendation ೥ऩਪఆ Salary Prediction ۀछɾ৬छਪఆ Job Category Prediction ٻਓಛ௃ਪఆ Prediction of Job Characteristics x ML
14. ٻ৬ऀͱٻਓͷϚονϯά Search Quality and Recommendation ೥ऩਪఆ Salary Prediction ۀछɾ৬छਪఆ Job Category Prediction ٻਓಛ௃ਪఆ Prediction of Job Characteristics x ML
15. ٻ৬ऀͱٻਓͷϚονϯά Search Quality and Recommendation ೥ऩਪఆ Salary Prediction ۀछɾ৬छਪఆ Job Category Prediction ٻਓಛ௃ਪఆ Prediction of Job Characteristics x ML
16. ٻ৬ऀͱٻਓͷϚονϯά Search Quality and Recommendation ೥ऩਪఆ Salary Prediction ۀछɾ৬छਪఆ Job Category Prediction ٻਓಛ௃ਪఆ Prediction of Job Characteristics x ML
17. ٻ৬ऀͱٻਓͷϚονϯά Search Quality and Recommendation ೥ऩਪఆ Salary Prediction ۀछɾ৬छਪఆ Job Category Prediction ٻਓಛ௃ਪఆ Prediction of Job Characteristics x ML
18. τϨʔχϯάσʔλ σʔλ͝ͱES΁ͷόονΠϯϙʔτεΫϦϓτΛ४උ͍ͯͨ͠
 ֶशͷ؀ڥ΍࣮ߦϑϩʔ͕୲౰ऀ͝ͱʹଐਓԽ͓ͯ͠Γɺ
 τϨʔχϯάσʔλͷϑΥʔϚοτ΋ఆ·͍ͬͯ·ͤΜɻ ֶशॲཧͷ࣮ߦ࣌ؒ σʔλྔͷ૿Ճʹͱ΋ͳ͏ֶशͷ௕࣌ؒԽ x MLͷ͓ͳ΍Έ
19. τϨʔχϯάσʔλ σʔλ͝ͱES΁ͷόονΠϯϙʔτεΫϦϓτΛ४උ͍ͯͨ͠
 ֶशͷ؀ڥ΍࣮ߦϑϩʔ͕୲౰ऀ͝ͱʹଐਓԽ͓ͯ͠Γɺ
 τϨʔχϯάσʔλͷϑΥʔϚοτ΋ఆ·͍ͬͯ·ͤΜɻ ֶशॲཧͷ࣮ߦ࣌ؒ σʔλྔͷ૿Ճʹͱ΋ͳ͏ֶशͷ௕࣌ؒԽ x MLͷ͓ͳ΍Έ
20. ֶशϞσϧͷετϨʔδ ػೳ͝ͱʹֶशϞσϧͷอଘઌ΋ଐਓԽ ༧ଌͷWeb API ػೳ͝ͱʹTornadoͳͲͰ؆қͳAPIαʔόΛ࡞੒͓ͯ͠Γɺ
 ͢͹΍͘ΞϓϦʹ൓өͰ͖ͳ͍ ༧ଌ݁ՌΛฦ͢APIͱͯ͠ESΛར༻͢ΔͳͲ͍ͯͨ͠ x MLͷ͓ͳ΍Έ
21. ֶशϞσϧͷετϨʔδ ػೳ͝ͱʹֶशϞσϧͷอଘઌ΋ଐਓԽ ༧ଌͷWeb API ػೳ͝ͱʹTornadoͳͲͰ؆қͳAPIαʔόΛ࡞੒͓ͯ͠Γɺ
 ͢͹΍͘ΞϓϦʹ൓өͰ͖ͳ͍ ༧ଌ݁ՌΛฦ͢APIͱͯ͠ESΛར༻͢ΔͳͲ͍ͯͨ͠ x MLͷ͓ͳ΍Έ
22. AIࣨͷ͓ͳ΍Έ શࣾͷ༷ʑͳࣄۀͱ࿈ܞ ෳ਺ͷࣄۀ͕͋ΓɺػցֶशͷػೳΛఏڙ͍ͯ͠Δ ֤ࣄۀͱσʔλ࿈ܞͷI/FΛڞ௨Խ͍ͨ͠ ։ൃϑϩʔΛڞ௨Խ͍ͨ͠ ઐ೚ͷΠϯϑϥΤϯδχΞ΋͍ͳ͍ ఏڙ͢Δػೳ͝ͱʹɺݸผʹΠϯϑϥΛ੔͑Δͱແବ͕ଟ͍ ग़དྷΔݶΓϑϨʔϜϫʔΫԽ͍ͨ͠
23. AIࣨͷ͓ͳ΍Έ શࣾͷ༷ʑͳࣄۀͱ࿈ܞ ෳ਺ͷࣄۀ͕͋ΓɺػցֶशͷػೳΛఏڙ͍ͯ͠Δ ֤ࣄۀͱσʔλ࿈ܞͷI/FΛڞ௨Խ͍ͨ͠ ։ൃϑϩʔΛڞ௨Խ͍ͨ͠ ઐ೚ͷΠϯϑϥΤϯδχΞ΋͍ͳ͍ ఏڙ͢Δػೳ͝ͱʹɺݸผʹΠϯϑϥΛ੔͑Δͱແବ͕ଟ͍ ग़དྷΔݶΓϑϨʔϜϫʔΫԽ͍ͨ͠
24. Solution ղܾࡦΛ୳͠·ͨ͠
26. 4
    3FETIJGU
    3%4ͷσʔλ͔Β
    ֶशϞσϧΛ࡞੒Ͱ͖Δ
    ༧ଌͷͨΊʹΫΤϦΛ࣮ߦͰ͖Δ ֶशϞσϧ͔Β༧ଌ஋Λฦ͢
    8FCΞϓϦέʔγϣϯΛࣗಈੜ੒
27. Solution
28. Solution Machine Leaning as a Service ML Tools
29. Solution Machine Leaning as a Service ML Tools Open Source
30. Machine Learning Stacks Apps Algorithm Processing Datastore API Server (Tornado…) Scikitlearn, SparkML … DL: Caffe2, DL4j, Tensorflow, Chainer … Hadoop, Spark, Storm … Elasticsearch, HBASE, Redshift …
31. Machine Learning Stacks Apps Algorithm Processing Datastore API Server (Tornado…) Scikitlearn, SparkML … DL: Caffe2, DL4j, Tensorflow, Chainer … Hadoop, Spark, Storm … Elasticsearch, HBASE, Redshift … PredictionIO
33. PredictionIO?
34. Salesforce Acquires PredictionIO
35. Salesforce Acquires PredictionIO Feb 19, 2016 - TechCrunch 4BMFTGPSDF
    ͕ػցֶशϓϥοτϑΥʔϜͷ
    1SFEJDUJPO*0Λങऩ
    
36. Salesforce Introduces Salesforce Einstein
37. Salesforce Introduces Salesforce Einstein Sep 18, 2016 - TechCrunch "*ΛऔΓࠐΉ4BMFTGPSDFͷ໺๬
    ػցֶशϓϥοτϑΥʔϜʮ&JOTUFJOʯΛൃද
38. The most stars repositories on Github? spark apache/spark ˒ 12.8k incubator-predictionio apache/incubator-predictionio ˒ 10.2k playframework playframework/playframework ˒ 9.3k scala scala/scala ˒ 8.2k
39. spark apache/spark ˒ 12.8k incubator-predictionio apache/incubator-predictionio ˒ 10.2k playframework playframework/playframework ˒ 9.3k scala scala/scala ˒ 8.2k The most stars repositories on Github? ˒10.2k
40. What is PredictionIO?
41. Apache PredictionIO? Apache PredictionIO (incubating) is an open source Machine Learning Server built on top of state-of-the-art open source stack for developers and data scientists create predictive engines for any machine learning task.
42. Apache PredictionIO (incubating) is an open source Machine Learning Server built on top of state-of-the-art open source stack for developers and data scientists create predictive engines for any machine learning task. Apache PredictionIO? ࠷ઌ୺ͷΦʔϓϯιʔεΛ
    ૊߹Θͤͨػցֶशαʔό
43. Apache PredictionIO (incubating) is an open source Machine Learning Server built on top of state-of-the-art open source stack for developers and data scientists create predictive engines for any machine learning task. Apache PredictionIO? ࠷ઌ୺ͷΦʔϓϯιʔεΛ
    ૊߹Θͤͨػցֶशαʔό ͲΜͳػցֶशλεΫͰ΋
    ༧ଌΤϯδϯ͕ͭ͘ΕΔ
44. Apache PredictionIO let you ର৅໰୊͝ͱʹςϯϓϨʔτΛ࡞Γɺ
 ͙͢ʹσϓϩΠͰ͖Δ quickly build and deploy an engine as a web service on production with customizable templates; ΫΤϦ౤͛ͯ݁ՌΛฦ͢API͕͋Δ respond to dynamic queries in real-time once deployed as a web service;
45. Apache PredictionIO let you ޡࠩͷௐ੔΍ɺධՁͷ࢓૊Έ΋͋Δ evaluate and tune multiple engine variants systematically; όον or ϦΞϧλΠϜͰ
 ֶशσʔλΛొ࿥͢ΔI/F͕͋Δ unify data from multiple platforms in batch or in real-time for comprehensive predictive analytics;
49. REST: EventAPI SDK: EventClient
51. Engine Template
53. Photo by Bernard Spragg. NZ Quick Start
54. Versions Latest Release Version v0.11.0
55. Quick Startʢ೔ຊޠʣ takezoe.hatenablog.com/entry/2017/05/11/132410
56. Photo by Bernard Spragg. NZ Algorithm
58. Photo by Bernard Spragg. NZ System Architecture
59. System Architecture Apache Hadoop up to 2.7.2 required only if YARN and HDFS are needed
 Apache HBase up to 1.2.4 Apache Spark up to 1.6.3
 for Hadoop 2.6 not Spark 2.x version Elasticsearch up to 1.7.5 not the Elasticsearch 2.x version
60. Storage roles Meta Data Event Data Model Data ✓ ✓ ✓ ✓ ✓* ✓ ✓ LOCALFS ✓
61. Photo by Bernard Spragg. NZ Implementation
62. Scala ੡ ɾ ػցֶशج൫ PredictionIOͱ SparkʹΑΔϨίϝϯυγ ες Ϝ
63. x ML
64. System Requirements ελϯόΠͷϨίϝϯυཁ݅ ϢʔβͷΫϦοΫϩά ͓ؾʹೖΓ௥Ճϩά S3ʹϩάσʔλ্͕͕͍ͬͯΔ ֶश͸೔࣍Ͱ
65. ElasticsearchͱTasteϓϥάΠϯͰ ࡞ΔϨίϝϯυγεςϜ
66. PIOಋೖલ σʔλ༻ͷESΠϯσοΫΛຖճ࡞੒ σʔλΠϯϙʔτ༻ͷઐ༻εΫϦϓτ Elasticsearch TasteϓϥάΠϯͰ࣮ߦ ศར͕ͩ൚༻ੑɺσʔλ૿ʹΑΔ࣮ߦ͕࣌ؒ՝୊ʹ ֶश݁ՌΛόϧΫϑΝΠϧͱͯ͠ग़ྗ Elasticsearch ༧ଌͷAPIͱͯ͠ར༻ ࣮ߦϑϩʔΛγΣϧεΫϦϓτͰ؅ཧ
67. PIOಋೖલ σʔλ༻ͷESΠϯσοΫΛຖճ࡞੒ σʔλΠϯϙʔτ༻ͷઐ༻εΫϦϓτ Elasticsearch TasteϓϥάΠϯͰ࣮ߦ ศར͕ͩ൚༻ੑɺσʔλ૿ʹΑΔ࣮ߦ͕࣌ؒ՝୊ʹ ֶश݁ՌΛόϧΫϑΝΠϧͱͯ͠ग़ྗ Elasticsearch ༧ଌͷAPIͱͯ͠ར༻ ࣮ߦϑϩʔΛγΣϧεΫϦϓτͰ؅ཧ
68. Click Log Favorite Log Event Server ALS Template
69. Click Log Favorite Log Event Server ALS Template pio import
70. Click Log Favorite Log Elasticsearch v5.3 cluster Event Server ALS Template pio import Data
71. Click Log Favorite Log Elasticsearch v5.3 cluster Event Server ALS Template pio import Data Spark 2 node cluster RDD
72. Click Log Favorite Log Elasticsearch v5.3 cluster Event Server ALS Template pio import Data LOCALFS Spark 2 node cluster RDD Model
73. Click Log Favorite Log Elasticsearch v5.3 cluster Event Server ALS Template pio import Data LOCALFS Spark 2 node cluster RDD Model Query Predicted Result
74. Engine Template?
75. Engine Template
76. D A S E D-A-S-E Data Source and Data Preparator Algorithm Serving Evaluation Metrics
77. Machine Learning Flow τϨʔχϯάσʔλ
    Training Data ػցֶशΞϧΰϦζϜ
    Machine Learning Algorithm ༧ଌϞσϧ
    Predictive Model લॲཧ
    Preprocessing Πϯϓοτσʔλ
    Input Data ༧ଌॲཧ
    Prediction Server ༧ଌ݁Ռ
    Predicted Result
78. Machine Learning Flow τϨʔχϯάσʔλ
    Training Data ػցֶशΞϧΰϦζϜ
    Machine Learning Algorithm ༧ଌϞσϧ
    Predictive Model લॲཧ
    Preprocessing Πϯϓοτσʔλ
    Input Data ༧ଌॲཧ
    Prediction Server ༧ଌ݁Ռ
    Predicted Result
79. Machine Learning Flow τϨʔχϯάσʔλ
    Training Data ػցֶशΞϧΰϦζϜ
    Machine Learning Algorithm ༧ଌϞσϧ
    Predictive Model લॲཧ
    Preprocessing Πϯϓοτσʔλ
    Input Data ༧ଌॲཧ
    Prediction Server ༧ଌ݁Ռ
    Predicted Result
80. Machine Learning Flow τϨʔχϯάσʔλ
    Training Data ػցֶशΞϧΰϦζϜ
    Machine Learning Algorithm ༧ଌϞσϧ
    Predictive Model લॲཧ
    Preprocessing Πϯϓοτσʔλ
    Input Data ༧ଌॲཧ
    Prediction Server ༧ଌ݁Ռ
    Predicted Result
81. Machine Learning Flow τϨʔχϯάσʔλ
    Training Data ػցֶशΞϧΰϦζϜ
    Machine Learning Algorithm ༧ଌϞσϧ
    Predictive Model લॲཧ
    Preprocessing Πϯϓοτσʔλ
    Input Data ༧ଌॲཧ
    Prediction Server ༧ଌ݁Ռ
    Predicted Result
82. Machine Learning Flow τϨʔχϯάσʔλ
    Training Data ػցֶशΞϧΰϦζϜ
    Machine Learning Algorithm ༧ଌϞσϧ
    Predictive Model લॲཧ
    Preprocessing Πϯϓοτσʔλ
    Input Data ༧ଌॲཧ
    Prediction Server ༧ଌ݁Ռ
    Predicted Result
83. Machine Learning Flow τϨʔχϯάσʔλ
    Training Data ػցֶशΞϧΰϦζϜ
    Machine Learning Algorithm ༧ଌϞσϧ
    Predictive Model લॲཧ
    Preprocessing Πϯϓοτσʔλ
    Input Data ༧ଌϞσϧ
    Predictive Model ༧ଌ݁Ռ
    Predicted Result Data Source & Preparator D
84. Machine Learning Flow τϨʔχϯάσʔλ
    Training Data ػցֶशΞϧΰϦζϜ
    Machine Learning Algorithm ༧ଌϞσϧ
    Predictive Model લॲཧ
    Preprocessing Πϯϓοτσʔλ
    Input Data ༧ଌϞσϧ
    Predictive Model ༧ଌ݁Ռ
    Predicted Result Data Source & Preparator D Algorithm A
85. Machine Learning Flow τϨʔχϯάσʔλ
    Training Data ػցֶशΞϧΰϦζϜ
    Machine Learning Algorithm ༧ଌϞσϧ
    Predictive Model લॲཧ
    Preprocessing Πϯϓοτσʔλ
    Input Data ༧ଌϞσϧ
    Predictive Model ༧ଌ݁Ռ
    Predicted Result Data Source & Preparator D Algorithm A Serving S
86. Machine Learning Flow τϨʔχϯάσʔλ
    Training Data ػցֶशΞϧΰϦζϜ
    Machine Learning Algorithm ༧ଌϞσϧ
    Predictive Model લॲཧ
    Preprocessing Πϯϓοτσʔλ
    Input Data ༧ଌϞσϧ
    Predictive Model ༧ଌ݁Ռ
    Predicted Result Data Source & Preparator D Algorithm A Serving S E Evaluation Metrics
88. D
89. D A
91. Machine Learning Flow τϨʔχϯάσʔλ
    Training Data ػցֶशΞϧΰϦζϜ
    Machine Learning Algorithm ༧ଌϞσϧ
    Predictive Model લॲཧ
    Preprocessing Πϯϓοτσʔλ
    Input Data ༧ଌϞσϧ
    Predictive Model ༧ଌ݁Ռ
    Predicted Result Data Source & Preparator D Algorithm A Serving S E Evaluation Metrics
92. D
93. DataSource •Event Store (Event Server) ͔ΒσʔλΛಡࠐ •TrainingDataΛฦ͢
96. D
97. Preparator • TrainingDataʹର͢Δલॲཧ • ಛ௃நग़ • ෳ਺AlgorithmΛར༻͢Δ৔߹ͷڞ௨ॲཧ • PreparedDataʹม׵ͯ͠Algoritmʹ౉͢
100. Machine Learning Flow τϨʔχϯάσʔλ
     Training Data ػցֶशΞϧΰϦζϜ
     Machine Learning Algorithm ༧ଌϞσϧ
     Predictive Model લॲཧ
     Preprocessing Πϯϓοτσʔλ
     Input Data ༧ଌϞσϧ
     Predictive Model ༧ଌ݁Ռ
     Predicted Result Data Source & Preparator D Algorithm A Serving S E Evaluation Metrics
101. A
102. Algorithm • train() Λ࣮૷ • ༧ଌϞσϧͷֶशΛ୲౰͢Δ • pio train ίϚϯυͰݺͼग़͞ΕΔ • HDFSʢLocalFSʣʹετΞ͞ΕΔ • predict() Λ࣮૷ • σϓϩΠޙͷΫΤϦʹରͯ͠ϦΞϧλΠϜʹݺ͹ΕΔ
108. Machine Learning Flow τϨʔχϯάσʔλ
     Training Data ػցֶशΞϧΰϦζϜ
     Machine Learning Algorithm ༧ଌϞσϧ
     Predictive Model લॲཧ
     Preprocessing Πϯϓοτσʔλ
     Input Data ༧ଌϞσϧ
     Predictive Model ༧ଌ݁Ռ
     Predicted Result Data Source & Preparator D Algorithm A Serving S E Evaluation Metrics
110. Serving • LServeΛܧঝ • serve() Λ࣮૷
112. Machine Learning Flow τϨʔχϯάσʔλ
     Training Data ػցֶशΞϧΰϦζϜ
     Machine Learning Algorithm ༧ଌϞσϧ
     Predictive Model લॲཧ
     Preprocessing Πϯϓοτσʔλ
     Input Data ༧ଌϞσϧ
     Predictive Model ༧ଌ݁Ռ
     Predicted Result Data Source & Preparator D Algorithm A Serving S E Evaluation Metrics
113. Precision@k Precision@5 / Threshold = 2.0 Predicted A ˒ˑˑ Validation B ˒˒˒ C ˒˒ˑ D ˑˑˑ E ˒˒ˑ A ˒ˑˑ B ˒˒˒ X ˒˒ˑ D ˑˑˑ E ˒˒ˑ
114. Precision@k Precision@5 / Threshold = 2.0 Predicted A ˒ˑˑ Validation B ˒˒˒ C ˒˒ˑ D ˑˑˑ E ˒˒ˑ A ˒ˑˑ B ˒˒˒ X ˒˒ˑ D ˑˑˑ E ˒˒ˑ
115. Precision@k Precision@5 / Threshold = 2.0 Predicted A ˒ˑˑ Validation B ˒˒˒ C ˒˒ˑ D ˑˑˑ E ˒˒ˑ A ˒ˑˑ B ˒˒˒ X ˒˒ˑ D ˑˑˑ E ˒˒ˑ PositiveCount: 2.0
116. x ML 5 Jobs
119. Photo by Bernard Spragg. NZ Conclusion
120. τϨʔχϯάσʔλ ϦΞϧλΠϜͰ΋ɺόονͰ΋σʔλΛऔΓࠐΉ͜ͱ͕Ͱ͖Δ ΞΫηετʔΫϯΛൃߦͰ͖ΔͷͰɺ֤αʔϏεͱͷ࿈ܞ͕ศར Elasticsearchͷ෼ࢄετϨʔδͷػೳΛڗडͰ͖Δ ֶशॲཧͷ࣮ߦ࣌ؒ SparkͷΫϥελΛ࢖͏ͨΊɺॲཧΛ෼ࢄֶ͠शʹ͔͔Δ࣌ؒΛ୹ॖ Open Source Machine Learning Server
121. ֶशϞσϧͷετϨʔδ ελϯόΠͰ͸LOCALFSΛར༻͍ͯ͠Δ
 ϞσϧͷಛੑʹԠͯ͡HDFSΛબ୒Մೳ ༧ଌͷWeb API “pio deploy” ίϚϯυ͚ͩͰ༧ଌͷAPIΛ࡞੒Ͱ͖Δ APIαʔό͸Akka-Httpϕʔε Open Source Machine Learning Server
122. ୤ɾଐਓԽ
123. Photo by Bernard Spragg. NZ Appendix
124. ϏζϦʔνͰ͸ػցֶशΞϓϦέʔγϣϯΛ
     ૊৫తʹ։ൃ͍ͯ͘͠ʹ͋ͨΓɺ։ൃɾӡ༻ج൫ͱ ͯ͠"QBDIF
     1SFEJDUJPO*0ʹऔΓ૊ΜͰ͍·͢ɻ
     ೥݄
     ฐ͔ࣾΒ໊͕ίϛολͱͯ͠ 1SFEJDUJPO*0ͷ։ൃʹࢀՃ͢Δ͜ͱʹͳΓ·ͨ͠ɻ Shinsuke Sugaya Naoki
 Takezoe Takako
 Shimamoto Takahiro Hagino
125. ϏζϦʔνͰ͸ػցֶशΞϓϦέʔγϣϯΛ
     ૊৫తʹ։ൃ͍ͯ͘͠ʹ͋ͨΓɺ։ൃɾӡ༻ج൫ͱ ͯ͠"QBDIF
     1SFEJDUJPO*0ʹऔΓ૊ΜͰ͍·͢ɻ
     ೥݄
     ฐ͔ࣾΒ໊͕ίϛολͱͯ͠ 1SFEJDUJPO*0ͷ։ൃʹࢀՃ͢Δ͜ͱʹͳΓ·ͨ͠ɻ Shinsuke Sugaya Naoki
 Takezoe Takako
 Shimamoto Takahiro Hagino
126. How to Contribute to PIO
127. Add support for Elasticsearch 5.x
129. jpioug.org
130. • ೔ຊ Apache PredictionIO Ϣʔβձ JPIOUG Join Us!
132. 30 6݄ FRI
133. 30 6݄ FRI Open Source Machine Learning Server 02 JPIOUG Meetup 19:30 @Shibuya
134. 30 6݄ FRI Open Source Machine Learning Server 02 JPIOUG Meetup 19:30 @Shibuya ʲٸืʳLT͍ͨ͠ํ
135. 30 8݄ WED
136. 30 8݄ WED Open Source Machine Learning Server 03 JPIOUG Meetup 19:30 @Shibuya
137. 30 8݄ WED Open Source Machine Learning Server 03 JPIOUG Meetup 19:30 @Shibuya ʲΏΔืʳLT͍ͨ͠ํ
138. jpioug.org
139. Open Source Machine Learning Server 15 Machine Learning minutes! Thank You