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
Build2019で発表された機械学習系をためしてみた
Search
Masakazu Muraoka
May 23, 2019
Technology
0
89
Build2019で発表された機械学習系をためしてみた
Masakazu Muraoka
May 23, 2019
Tweet
Share
More Decks by Masakazu Muraoka
See All by Masakazu Muraoka
たぶんオレのはPMじゃない
bathtimefish
0
34
猫と私とGCP
bathtimefish
4
540
Other Decks in Technology
See All in Technology
Introducing RFC9111 / YAPC::Fukuoka 2025
k1low
1
210
QAセントラル組織が運営する自動テストプラットフォームの課題と現状
lycorptech_jp
PRO
0
350
Amazon ECS デプロイツール ecspresso の開発を支える「正しい抽象化」の探求 / YAPC::Fukuoka 2025
fujiwara3
8
1.3k
「データ無い! 腹立つ! 推論する!」から 「データ無い! 腹立つ! データを作る」へ チームでデータを作り、育てられるようにするまで / How can we create, use, and maintain data ourselves?
moznion
6
2.9k
AIと共に開発する時代の組織、プロセス設計 freeeでの実践から見えてきたこと
freee
3
570
Design and implementation of "Markdown to Google Slides" / phpconfuk 2025
k1low
1
390
AWS資格は取ったけどIAMロールを腹落ちできてなかったので、年内に整理してみた
hiro_eng_
0
170
エンジニアにとってコードと並んで重要な「データ」のお話 - データが動くとコードが見える:関数型=データフロー入門
ismk
0
450
Datadog On-Call と Cloud SIEM で作る SOC 基盤
kuriyosh
0
150
“それなりに”安全なWebアプリケーションの作り方
xryuseix
0
270
内部品質・フロー効率・コミュニケーションコストを悪化させ現場を苦しめかねない16の組織設計アンチパターン[超簡易版] / 16 Organization Design Anti-Patterns for Software Development
mtx2s
2
190
決済システムの信頼性を支える技術と運用の実践
ykagano
0
450
Featured
See All Featured
Rebuilding a faster, lazier Slack
samanthasiow
84
9.3k
Designing Experiences People Love
moore
142
24k
Optimising Largest Contentful Paint
csswizardry
37
3.5k
Side Projects
sachag
455
43k
Templates, Plugins, & Blocks: Oh My! Creating the theme that thinks of everything
marktimemedia
31
2.6k
Design and Strategy: How to Deal with People Who Don’t "Get" Design
morganepeng
132
19k
"I'm Feeling Lucky" - Building Great Search Experiences for Today's Users (#IAC19)
danielanewman
231
22k
Large-scale JavaScript Application Architecture
addyosmani
514
110k
How Fast Is Fast Enough? [PerfNow 2025]
tammyeverts
3
310
Java REST API Framework Comparison - PWX 2021
mraible
34
9k
Helping Users Find Their Own Way: Creating Modern Search Experiences
danielanewman
31
2.9k
Learning to Love Humans: Emotional Interface Design
aarron
274
41k
Transcript
Copyright(c) Kobe Digital Labo Inc. #VJMEͰൃද͞ΕͨػցֶशܥΛͨΊͯ͠Έͨ ଜԬਖ਼
Copyright(c) Kobe Digital Labo Inc. HTML5-WEST.jpද / html5j ϚʔΫΞοϓ෦ ෦
/ HTML5 Experts.jp ϝϯόʔ NPO๏ਓຊΣΞϥϒϧσόΠεϢʔβʔձཧࣄ ਆށࢢΣΞϥϒϧσόΠεਪਐձٞϝϯόʔ JS Boardษڧձ ओ࠻ ΉΒ͓͔ɹ·͔ͣ͞ ଜԬਖ਼ גࣜձࣾਆށσδλϧɾϥϘ औక @bathtimefish 8FC *P5ؔ࿈ٕज़ʹ͍ͭͯͷߨԋࣥචΛ ΘΓͱͨ͘͞Μͬͯ·͢ɻ
#VJMEͰػցֶशؔ࿈ͷൃද͕͍͔ͭ͋ͬͨ͘ IUUQTXXXJUNFEJBDPKQFOUFSQSJTFBSUJDMFTOFXTIUNM IUUQTKBQBO[EOFUDPNBSUJDMF ग़ॴ;%/FU+BQBO ग़ॴ*5.FEJB
ͦͷதͰؾʹͳ͍͔ͬͨͭ͘ "[VSF1FSTPOBMJ[FS 1SFWJFX IUUQTB[VSFNJDSPTPGUDPNKBKQTFSWJDFTDPHOJUJWFTFSWJDFTQFSTPOBMJ[FS .-OFU"VUP.- 1SFWJFX IUUQTHJUIVCDPNEPUOFUNBDIJOFMFBSOJOHTBNQMFTCMPCNBTUFS 3&"%.&NEBVUPNBUFNMOFUNPEFMTHFOFSBUJPOQSFWJFXTUBUF "OPNBMZ%FUFDUPS 1SFWJFX
IUUQTB[VSFNJDSPTPGUDPNKBKQTFSWJDFTDPHOJUJWFTFSWJDFTQFSTPOBMJ[FS $SFBUFBOEFYQMPSFBVUPNBUFENBDIJOFMFBSOJOH FYQFSJNFOUTJOUIF"[VSFQPSUBM 1SFWJFX IUUQTEPDTNJDSPTPGUDPNFOVTB[VSFNBDIJOFMFBSOJOHTFSWJDFIPXUPDSFBUFQPSUBMFYQFSJNFOUT ✔ ✔ ͭ͞Βͬͱಈ͔ͯ͠ΈΑ͏ͱࢥ͍·͢
ಈ࡞ڥ NBD04W 7JTVBM4UVEJPGPS.BD$PNNVOJUZ CVJME /&5$PSF
"[VSF1FSTPOBMJ[FS 1SFWJFX IUUQTB[VSFNJDSPTPGUDPNKBKQTFSWJDFTDPHOJUJWFTFSWJDFTQFSTPOBMJ[FS
IUUQTEPDTNJDSPTPGUDPNFOVTB[VSFDPHOJUJWFTFSWJDFTQFSTPOBMJ[FSDTIBSQRVJDLTUBSUDPNNBOEMJOFGFFECBDLMPPQ
1PSUBMͰ1FSTPOBMJ[FSΛ σϓϩΠ͢Δ
%PDLFSDPOUBJOFS͕͋Δ͡ΌΜ͜ΕָͪΜʂ
ӕΛ͔ͭΕ·ͨ͠ IUUQTIVCEPDLFSDPN@NJDSPTPGUB[VSFDPHOJUJWFTFSWJDFTQFSTPOBMJ[FS
ΤϯυϙΠϯτͱΩʔΛϝϞ͓ͬͯ͘
Ϟσϧͷߋ৽සΛසൟʹ͓ͯ͘͠
7JTVBM4UVEJPͰ/&5ίϯιʔϧΞϓϦΛ࡞͢Δ
cd ./PersonalizerExample/PersonalizerExample dotnet add package Microsoft.Azure.CognitiveServices.Personalizer --version 0.8.0-preview 1FSTPOBMJ[FS&YBNQMFDTQSPK͕͋ΔσΟϨΫτϦʹҠಈͯ͠ 1FSTPOBMJ[FS"1*ύοέʔδΛՃ͢Δ
IUUQTEPDTNJDSPTPGUDPNFOVTB[VSFDPHOJUJWFTFSWJDFTQFSTPOBMJ[FS DTIBSQRVJDLTUBSUDPNNBOEMJOFGFFECBDLMPPQBEEDPEFUPSBOLUIFBDUJPOT ZPVXBOUUPTIPXUPZPVSVTFST ͷ1SPHSBNDTΛίϐϖͯ͠ ΤϯυϙΠϯτͱΩʔΛઃఆ͢Δ
Ϗϧυ͢ΔͱΞϓϦ͕ىಈ͢Δ
ྉཧΛ͓͢͢Ίͯ͘͠ΔͷͰɺ:FT/PͰΈΛճ͍ͯ͘͠
͠Β͘͢ΔͱQSPCBCJMJUZͷείΞ͕มಈͨ͠ ͳΜֶ͔शͯ͠ΔͬΆ͍
Ϟσϧͷ࠷ऴߋ৽͕ఆظతʹߋ৽͞Ε͍ͯΔ
ΞΫγϣϯʹԠͨ͡ϦϫʔυϥϯΩϯάΛऔಘͯ͠ ճʹԠͨ͡ϦϫʔυΛઃఆͯ͠ϦΫΤετ͢Δ
IUUQTHJUIVCDPNEPUOFUNBDIJOFMFBSOJOHTBNQMFTCMPCNBTUFS3&"%.&NEBVUPNBUFNMOFUNPEFMTHFOFSBUJPOQSFWJFXTUBUF .-/&5Ͱ"VUP.-"1*͕αϙʔτ͞Εͨ
IUUQTFOXJLJQFEJBPSHXJLJ"VUPNBUFE@NBDIJOF@MFBSOJOH
IUUQTHJUIVCDPNEPUOFUNBDIJOFMFBSOJOHTBNQMFTUSFFNBTUFSTBNQMFTDTIBSQHFUUJOHTUBSUFE3FHSFTTJPO@"VUP.- λΫγʔͷྉۚ༧ଌͷΛͬͯΈΔ
git clone https://github.com/dotnet/machinelearning-samples.git cd ./machinelearning-samples/samples/csharp/getting-started/Regression_AutoML $MPOFͯ͠αϯϓϧ·ͰҠಈ͢Δ
brew install plplot 1-1MPUΛ͏ͷͰΠϯετʔϧ͢Δ
- string chartFileNamePath = @".\" + chartFileName; + string chartFileNamePath
= chartFileName; ߦΛҎԼͷΑ͏ʹमਖ਼͢Δ मਖ਼͠ͳ͍ͱධՁάϥϑ͕/PUGPVOEʹͳͬͯ։͔ͳ͍ ͳΜͰΘ͟Θ͟ϑΝΠϧ໊ͷઌ಄ʹaΛ͚͍ͭͯΔͷ͔Θ͔Βͳ͍ 8JOEPXTͩͱ=͔ʁ
Ϗϧυ࣮ͯ͠ߦ͢Δ
͍Ζ͍ΖͳճؼੳΛࢼͯ͠ϕετͳϞσϧΛબ͍ͯ͠Δ
ͦΕͳΓͷਫ਼͕Ͱͨ
ΩϞ͜ͷߦ -BCFM$PMVNO/BNFʹλʔήοτΧϥϜͷϥϕϧ໊ 'BSF"NPVOU Λࢦఆͯ͠ $SFBUF3FHSFTTJPO&YQFSJNFOUͰճؼੳͷ࣮ݧثΛ࡞͠ ࢦఆ࣌ؒ ඵ ؒ͞·͟·ͳճؼੳΛࢼߦͯ͠#FTU.PEFMΛબͼग़͢ #JOBSZ$MBTTJpDBUJPOɺ.VMUJDMBTT$MBTTJpDBUJPOಉ༷ͷίʔυελΠϧ
·ͱΊ "VUP.-͕खܰʹ͑ΔΑ͏ʹͳͬͨ͜ͱͰɺϩʔΧϧͰֶशਪ͢Δ͜ͱ ͕༰қʹͳͬͨɻϩʔΧϧʹஷΊͨσʔλΛࣗಈతʹֶशͯ͠ਫ਼͕ ্͕Δ"*తػೳΛΞϓϦʹ࣋ͨͤΔ͜ͱ͕؆୯ʹͳΔ͔͠Εͳ͍ Ұ෦ΤϥʔΛు͘αϯϓϧίʔυ͚͋ͬͨͲͦͷ͏ͪΔʹ͕͍ͪͳ͍ $͕͍͍ײ͡ʹΫϩεϓϥοτϑΥʔϜݴޠʹͳ͖͔ͬͯͨΒ.-OFUͷ৳ͼ͠Ζʹظ Ϋϥυͷ"VUP.-($1ͳΜ͔͕ڧ͍͚Ͳׂͱߴ͍ɻͬͪࣗ͜Ͱॻ͚λμɻ "VUP.-ͳ'SBNFXPSLʹ"VUP,FSBT IUUQTBVUPLFSBTDPN ͕͋Δ͚Ͳ
ϗϫΠτΧϥʔࣾசʹ/&5ͷ΄͏͕͍͍Μ͡ΌͶʁ (16TVQQPSUΑ IUUQTEFWCMPHTNJDSPTPGUDPNEPUOFU BOOPVODJOHNMOFUNBDIJOFMFBSOJOHGPSOFU
Thanks !