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
Introduction of Deep Learning for Rubyist
Search
tanaken0515
November 30, 2019
Programming
2
2.4k
Introduction of Deep Learning for Rubyist
tanaken0515
November 30, 2019
Tweet
Share
More Decks by tanaken0515
See All by tanaken0515
Lightning Talks | Kaigi on Rails _2022_ new
tanaken0515
1
4.1k
Organization for Pepabo Customer Engineers
tanaken0515
0
380
Proud of SUZURI CRE Team
tanaken0515
0
1.8k
CRE as PEPABO
tanaken0515
0
870
図解:「ソフトウェアの仕様、挙動、バグ、利用者の期待」
tanaken0515
2
500
Retrospective of SUZURI CRE 1st year
tanaken0515
0
1.8k
Creator Mind of SUZURI
tanaken0515
2
1.3k
Rails Girls Tokyo 13th Sponsor Lightning Talk
tanaken0515
0
320
Other Decks in Programming
See All in Programming
WordPress Playground for Developers
iambherulal
0
120
家族・子育て重視/沖縄在住を維持しながらエンジニアとしてのキャリアをどのように育てていくか?
ug
0
240
生産性アップのためのAI個人活用
kunoyasu
0
650
CRE Meetup!ユーザー信頼性を支えるエンジニアリング実践例の発表資料です
tmnb
0
360
Compose Navigation実装の見通しを良くする
hiroaki404
0
180
Kubernetesで実現できるPlatform Engineering の現在地
nwiizo
2
1.7k
AtCoder Heuristic First-step Vol.1 講義スライド(山登り法・焼きなまし法編)
takumi152
3
990
生成AIの使いどころ
kanayannet
0
100
ミリしらMCP勉強会
watany
3
440
Return of the Full-Stack Developer
simas
PRO
1
320
RCPと宣言型ポリシーについてのお話し
kokitamura
2
150
remix + cloudflare workers (DO) docker上でいい感じに開発する
yoshidatomoaki
0
120
Featured
See All Featured
Scaling GitHub
holman
459
140k
Side Projects
sachag
452
42k
The Illustrated Children's Guide to Kubernetes
chrisshort
48
49k
Intergalactic Javascript Robots from Outer Space
tanoku
270
27k
Six Lessons from altMBA
skipperchong
27
3.7k
Understanding Cognitive Biases in Performance Measurement
bluesmoon
28
1.6k
No one is an island. Learnings from fostering a developers community.
thoeni
21
3.2k
Practical Tips for Bootstrapping Information Extraction Pipelines
honnibal
PRO
17
1.1k
A designer walks into a library…
pauljervisheath
205
24k
The Language of Interfaces
destraynor
157
24k
VelocityConf: Rendering Performance Case Studies
addyosmani
328
24k
A Philosophy of Restraint
colly
203
16k
Transcript
ʮ3VCZJTUͷͨΊͷ%FFQ-FBSOJOHೖʯฤ ాத݈ଠ(.01FQBCP *OD ࣛࣇౡ3VCZձٞ ͔ͬ͝·หͷ%FFQ-FBSOJOHΛ 3VCZͰ͖ͬ
ࣛࣇౡ3VCZձٞӡӦҕһ (.0ϖύϘגࣜձࣾ46;63*ࣄۀ෦ΤϯδχΞ ాத݈ଠ!UBOBLFO
ࣛࣇౡหͷػց༁Λ 3VCZͰΓ͍ͨ
ػց༁ͷΞϓϩʔν ϧʔϧϕʔεػց༁ 3VMF#BTFE.BDIJOF5SBOTMBUJPO ౷ܭతػց༁ 4UBUJTUJDBM#BTF.BDIJOF5SBOTMBUJPO χϡʔϥϧػց༁ /FVSBM.BDIJOF5SBOTMBUJPO (PPHMF༁ʹΘΕ͍ͯΔχϡʔϥϧػց༁ΛͬͯΈ͍ͨ
(PPHMF`T/FVSBM.BDIJOF5SBOTMBUJPO4ZTUFN#SJEHJOHUIF(BQCFUXFFO)VNBOBOE.BDIJOF5SBOTMBUJPO
dࣛࣇౡห༁ͷಓd 3VCZJTUͷͨΊͷ%FFQ-FBSOJOHೖ χϡʔϥϧػց༁ೖʙຊޠΛӳޠʹ༁͢Δʙ ࣛࣇౡห༁ʙํݴ༁ͷΞϓϩʔνʙ
dࣛࣇౡห༁ͷಓd 3VCZJTUͷͨΊͷ%FFQ-FBSOJOHೖ χϡʔϥϧػց༁ೖʙຊޠΛӳޠʹ༁͢Δʙ ࣛࣇౡห༁ʙํݴ༁ͷΞϓϩʔνʙ ࠓओʹ͜Εʹ͍ͭͯ͠·͢
dࣛࣇౡห༁ͷಓd 3VCZJTUͷͨΊͷ%FFQ-FBSOJOHೖ χϡʔϥϧػց༁ೖʙຊޠΛӳޠʹ༁͢Δʙ ࣛࣇౡห༁ʙํݴ༁ͷΞϓϩʔνʙ ͜ΕΒʹ͍ͭͯগ͚ͩ͠͠·͢
ࠓճͷΰʔϧ #FGPSF !ʮ%FFQ-FBSOJOHͬͨ͜ͱͳ͍ͳ͊ʯ ʮΔͱͯ͠1ZUIPO͡Όͳ͍ʁʯ "GUFS #ʮ3VCZͰ%FFQ-FBSOJOHΓ͍ͨʂʯ
͘͡ 3VCZͰ%FFQ-FBSOJOHΛΔʹʁ ۩ମతʹͲ͏࣮͢Δͷʁ %FFQ-FBSOJOHΛֶͿʹʁ
3VCZͰ %FFQ-FBSOJOHΛΔʹʁ
1ZUIPOͷ߹ ϑϨʔϜϫʔΫ $IBJOFS 5FOTPS'MPX .9/FU ܭࢉɺߦྻܭࢉ /VN1Z $V1Z IUUQTTQFBLFSEFDLDPNTPOPUTSFEDIBJOFS BOEDVNPQSBDUJDBMEFFQMFBSOJOHJOSVCZBU
SVCZLBJHJ TMJEF
3VCZͷ߹ ϑϨʔϜϫʔΫ 3FE$IBJOFS 5FOTPS'MPXSC .9/FUSC NFOPISVCZ ܭࢉɺߦྻܭࢉ /VNP/"SSBZ $VNP IUUQTTQFBLFSEFDLDPNTPOPUTSFEDIBJOFS
BOEDVNPQSBDUJDBMEFFQMFBSOJOHJOSVCZBU SVCZLBJHJ TMJEF
3VCZͰ%FFQ-FBSOJOH 3VCZ༻ͷϑϨʔϜϫʔΫ͕͋Δʂ ͋Δ͍ɺ1Z$BMMͱ͍͏SVCZHFNΛͬͯ 3VCZ͔Β1ZUIPOͷϥΠϒϥϦΛ͏͜ͱՄೳ
ͳΔ΄ͲͶ Ͱɺ۩ମతʹͲ͏࣮͢Δͷʁ
3FE$IBJOFS 1ZUIPOͷ$IBJOFSΛ3VCZͰҠ২ͨ͠ͷ 3VCZ༻ͷχϡʔϥϧωοτϫʔΫͷϑϨʔϜϫʔΫ 3VCZ,BJHJͷొஃ༰ʹ 3FE$IBJOFSBOE$VNP1SBDUJDBM%FFQ-FBSOJOHJO3VCZBU3VCZ,BJHJ4QFBLFS%FDL 3VCZGPS/-13VCZ,BJHJ %FFQ-FBSOJOH1SPHSBNNJOHPO3VCZ4QFBLFS%FDL
ྫΞϠϝ *SJT ͷछΛྨ͢Δ *SJT%BUB4FUϨίʔυ छYϨίʔυ ͕͘ยͷ͞ ͕͘ยͷ෯ ՖͼΒͷ͞ ՖͼΒͷ෯ ΞϠϝͷछ
*SJT4FUPTB *SJT7FSTJDPMPVS *SJT7JSHJOJDB σʔλΛֶशͯ͠छΛਪ͢Δ *OQVU͕͘หͷ͞ɾ෯ɺՖͼΒͷ͞ɾ෯ 0VUQVUछ
ਪͱֶश
ਪͷΠϝʔδ ԿΒ͔ͷॲཧ ೖྗ
ग़ྗ ֤छʹ ଐ͢Δ֬
.-1.VMUJMBZFS1FSDFQUSPO
.-1.VMUJMBZFS1FSDFQUSPO தؒ ·ͨɺӅΕ
.-1.VMUJMBZFS1FSDFQUSPO ઢܗม -
ߦ/ྻ ઢܗม - /ߦ/ྻ ઢܗม - /ߦྻ ઢܗมΛܦͯग़ྗ͢Δ /ӅΕͷϊʔυ
.-1.VMUJMBZFS1FSDFQUSPO தؒͰ ඇઢܗมߦͳ͏
ඇઢܗม 3F-V ඇઢܗม 3F-V
<࣮>3FE$IBJOFSͰ.-1 ֤Ͱߦ͏มΛఆٛ ࠓճͭͷઢܗม ೖྗY͕༩͑ΒΕͨ࣌ͷਪॲཧ ઢܗ ඇઢܗͷมΛ܁Γฦ͢
ਪ͜͜·Ͱɹଓֶ͍ͯश
ֶशͷΠϝʔδ ೖྗ ਪͱਖ਼ղΛൺͯΈΔͱʜʁ
ग़ྗ
ֶशͷΠϝʔδ ೖྗ ग़ྗ
ਖ਼ղ ਪͱਖ਼ղΛൺͯΈΔͱʜʁ
ֶशͷΠϝʔδ ೖྗ ग़ྗ
ਖ਼ղ ޡ͕ࠩ͋Δ ਪͱਖ਼ղΛൺͯΈΔͱʜʁ
ֶशͷΠϝʔδ ೖྗ ઢܗม - ߦ/ྻ ઢܗม
- /ߦ/ྻ ઢܗม - /ߦྻ ग़ྗ ਖ਼ղ
ֶशͷΠϝʔδ ೖྗ ઢܗม - ߦ/ྻ ઢܗม
- /ߦ/ྻ ઢܗม - /ߦྻ ग़ྗ ਖ਼ղ ޡ͕ࠩখ͘͞ͳΔΑ͏ʹ ઢܗมͷॏΈΛ ߋ৽͢Δ ֶश
ʮޡ͕ࠩখ͘͞ͳΔΑ͏ʹʯͱʁ ޡࠩΛଛࣦؔ -PTT ͱͯ͠ఆٛ͢Δ ͦΕͧΕͷॏΈύϥϝʔλ 8 ʹؔͯ͠ɺଛࣦؔͷޯΛٻΊΔ ޯ㲈͖ภඍͰٻΊΔ͜ͱ͕Ͱ͖Δ ॏΈύϥϝʔλΛޯͷٯํʹߋ৽͢Δޯ߱Լ๏ ྫ
֬తޯ߱Լ๏ 4%(4UPDIBTUJD(SBEJFOU%FTDFOU
ֶशΛ3FE$IBJOFSͰͬͯΈΔ
<࣮>σʔληοτͷ४උͱׂ ݸͷσʔλ <<Y Y Y Y> Z> <<Y Y Y
Y> Z> ʜ ֶश༻ ݕূ༻ ςετ༻ ֶश༻ͷσʔλΛ σʔλֶͣͭश͢Δ
<࣮>ਪͱֶशͷϩδοΫࢦఆ ਪ༻ͷωοτϫʔΫΛ࡞Δ ࠓճ.-1Λ͏
<࣮>ਪͱֶशͷϩδοΫࢦఆ ྨ༻ͷϥούʔ ࢦఆ͞ΕͨଛࣦؔΛͬͯଛࣦΛฦ͢Α͏ʹQSFEJDUPSΛϥοϓ͢Δ ະࢦఆͷ߹ަࠩΤϯτϩϐʔޡࠩ
<࣮>ਪͱֶशͷϩδοΫࢦఆ ॏΈΛߋ৽ ֶश ͢ΔϩδοΫΛࢦఆ PQUJNJ[FSʹج͍ͮͯɺֶश༻σʔλ ΠςϨʔλ Ͱ ਪଛࣦͷࢉग़ॏΈͷߋ৽ΛΔPCKFDU
<࣮>ֶशͷ࣮ߦ VQEBUFSʹΑΔֶशΛ TUPQ@USJHHFS·Ͱ܁Γฦ͢ ࠓճFQPDI ੈ
ֶशͷ༷ࢠ ଛࣦ͕ݮ͍͖ͬͯɺ ਖ਼ղ্͕͕͍ͬͯ͘
ֶशͷ༷ࢠ ݕূ༻σʔλ Ͱ ࣅͨ݁Ռͱͳͬͨ
ֶशͨ͠ޙ
<࣮>εφοϓγϣοτΛͬͨਪ ֶश࣌ͱಉ͡ߏͷωοτϫʔΫʹ εφοϓγϣοτΛ-PBE͢Δ ςετ༻σʔλͷਪ݁ՌΛ ਖ਼ղͱൺֱ͢Δ
ਪ݁Ռ ਖ਼ղ ෆਖ਼ղ݅
Θ͔͖ͬͯͨʂ ͬͱֶͿʹʁ
$IBJOFS5VUPSJBM͕Φεεϝ ४උฤ ͡Ίʹ 1ZUIPOೖ ػցֶशʹΘΕΔֶ ඍͷجૅ ઢܗͷجૅ ֬ɾ౷ܭͷجૅ ػցֶशͱσʔλੳೖ ୯ճؼੳͱॏճؼੳ
/VN1Zೖ TDJLJUMFBSOೖ $V1Zೖ 1BOEBTೖ .BUQMPUMJCೖ σΟʔϓϥʔχϯάೖ χϡʔϥϧωοτϫʔΫ ͷجૅ $IBJOFSͷجૅ $IBJOFSͷԠ༻ τϨʔφͱ ΤΫεςϯγϣϯ ແྉͰχϡʔϥϧωοτϫʔΫ·ͰͷجૅΛຬวͳֶ͘Δ
·ͱΊ 3VCZͰ%FFQ-FBSOJOHΛΔʹʁ 3VCZ༻ͷϑϨʔϜϫʔΫ͔ɺ1Z$BMMΛ͏ ۩ମతʹͲ͏࣮͢Δͷʁ 3FE$IBJOFSΛͬͨྨͷ࣮ྫΛհͨ͠ %FFQ-FBSOJOHΛֶͿʹʁ $IBJOFS5VUPSJBM͕Φεεϝ
3VCZͰ%FFQ-FBSOJOH Γͨ͘ͳ͖ͬͯͨਓ#ʁ
None
<༧ࠂ>ʮχϡʔϥϧػց༁ೖʯฤ ୯ޠͷϕΫτϧԽʂʁࢄදݱʁʁ ܗଶૉղੳ.F$BCʂHFN/BUUPʁ ࣗવݴޠॲཧͰ༻͍ΒΕΔχϡʔϥϧωοτϫʔΫͱʜʂʁ XPSEWFD $#08 DPOUJOVPVTCBHPGXPSET Ϟσϧ 3// SFDVSSFOUOFVSBMOFUXPSL
-45. MPOHTIPSUUFSNNFNPSZ 4FRVFODFUP4FRVFODF 3FE$IBJOFSͰͲ͜·Ͱ࣮ݱͰ͖Δͷ͔ʂʁ
<༧ࠂ>ʮࣛࣇౡห༁ʯฤ ํݴσʔλͲ͜ͰखʹೖΕΔʂʁ ώϯτࠃཱࠃޠݚڀॴʰํݴஊࢿྉʱʹ͋ͬͨʂ σʔλͷෆΛิ͏ʹʜʂʁ ࣅ͍ͯΔํݴʹண͢Δͱʂʁ ӳޠ༁Λͦͷ··ͯΊΔ͜ͱͰ͖ͳ͍ʂʁʂʁ ํݴͷςΩετɺൃԻΛ࣮ʹදݱ͢Δඞཁ͕͋Δ ୯ޠΑΓԻ͕େࣄͻΒ͕ͳɺΧλΧφɺΞϧϑΝϕοτ
3VCZͰࣛࣇౡห༁ Γͨ͘ͳ͖ͬͯͨਓ#ʁ
None
͍͕͋ͱ͛ͨ͞͠ 5IBOLZPVGPSMJTUFOJOH