Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Amazon SageMakerの最新アップデートの紹介

Sponsored · Ship Features Fearlessly Turn features on and off without deploys. Used by thousands of Ruby developers. →
Avatar for tandfy tandfy
December 16, 2019
Programming
960
1

Amazon SageMakerの最新アップデートの紹介

re:Growth 2019 Osakaでの発表内容です。

Avatar for tandfy

tandfy

December 16, 2019

More Decks by tandfy

See All by tandfy
アプリから集まるイベントデータのリアルタイム処理入門
Avatar for tandfy tandfy
0
1.7k
Amazon Personalizeではじめるレコメンドサービス
Avatar for tandfy tandfy
0
1.2k
DeepRacerで学ぶ機械学習 1.1
Avatar for tandfy tandfy
0
1.1k
DeepRacerで始める機械学習
Avatar for tandfy tandfy
1
1.8k
DeepRacerでまなぶ強化学習
Avatar for tandfy tandfy
1
1.5k
Amazon SageMakerではじめる物体検出
Avatar for tandfy tandfy
1
1.1k

Other Decks in Programming

See All in Programming
RailsTokyo 2026#4: AI様があれば、
Hotwireの弱点は消えるか？
Avatar for Naofumi Kagami naofumi
5
1k
[KCD Czech] eBPF Meets the GPU: Future of AI Infra Observability
Avatar for Donia Chaiehloudj doniacld
0
130
DynamoDBには集計系のクエリがないけどなんとかしたい
Avatar for 村上優稀 musan
1
110
AIエージェントと協働するCLI開発 — BunとOpenClawで学んだこと
Avatar for よしこ yoshikouki
1
230
運用エージェントは "作る" から "育てる" へ - 記憶と自己進化の3層設計パターン / self-evolving-agents-three-layer-agent-design
Avatar for geeawa gawa
12
3.3k
CLIであることを活かしたGitHub Copilot CLI活用術 / GitHub Copilot CLI Pro Tips & Tricks
Avatar for Naoya.i nao_mk2
1
1.2k
誰も頼んでない機能を出荷した話
Avatar for Hidetaka Toriyama zekutax
0
150
初めてのRubyKaigiはこう見えた
Avatar for asano jellyfish700
0
380
net-httpのHTTP/2対応について
Avatar for NARUSE, Yui naruse
0
390
AIとRubyの静的型付け
Avatar for ukin0k0 ukin0k0
0
490
Copilot CLI の継戦能力を高める コンテキスト管理
Avatar for Nozomuts nozomutu
1
1.1k
気づいたらRubyで100作品 ー クリエイティブコーディングが生活の一部になるまで / 100 Ruby Sketches Later: How Creative Coding Became Part of My Life
Avatar for chobishiba chobishiba
3
510

Featured

See All Featured
Beyond borders and beyond the search box: How to win the global "messy middle" with AI-driven SEO
Avatar for David Carrasco davidcarrasco
3
150
Claude Code のすすめ
Avatar for schroneko schroneko
67
220k
The State of eCommerce SEO: How to Win in Today's Products SERPs - #SEOweek
Avatar for Aleyda Solis aleyda
2
11k
Ecommerce SEO: The Keys for Success Now & Beyond - #SERPConf2024
Avatar for Aleyda Solis aleyda
1
2k
Why Your Marketing Sucks and What You Can Do About It - Sophie Logan
Avatar for Sophie Logan marketingsoph
0
160
For a Future-Friendly Web
Avatar for Brad Frost brad_frost
183
10k
Easily Structure & Communicate Ideas using Wireframe
Avatar for Afnizar Nur Ghifari afnizarnur
194
17k
Kristin Tynski - Automating Marketing Tasks With AI
Avatar for Tech SEO Connect techseoconnect
PRO
0
260
技術選定の審美眼（2025年版） / Understanding the Spiral of Technologies 2025 edition
Avatar for Takuto Wada twada
PRO
118
120k
It's Worth the Effort
Avatar for 3n 3n
188
29k
The World Runs on Bad Software
Avatar for Brandon Keepers bkeepers
PRO
72
12k
Leveraging LLMs for student feedback in introductory data science courses - posit::conf(2025)
Avatar for Mine Cetinkaya-Rundel minecr
1
270

Transcript

  1. "NB[PO4BHF.BLFS࠷৽Ξοϓσʔτ  େᖒ༐ే SF(SPXUI04","

  2. ࣗݾ঺հ  େᖒ༐ే σʔλΞφϦςΟΫεࣄۀຊ෦ΠϯςάϨʔγϣϯ෦ ػցֶशνʔϜΤϯδχΞ!େࡕ w 4BHF.BLFSؔ࿈ϒϩάຊ d  w

    SF*OWFOUࢀՃճ໨

  3. εϥΠυ͸ޙͰೖख͢Δ͜ͱ͕ग़དྷ·͢ͷͰ ൃදதͷ಺༰ΛϝϞ͢Δඞཁ͸͋Γ·ͤΜɻ ࣸਅࡱӨΛ͢Δ৔߹͸ ϑϥογϡɾγϟολʔԻ͕ग़ͳ͍Α͏ʹ͝഑ྀ͍ͩ͘͞ Attention

  4.  SF*OWFOU ػցֶशؔ࿈ͷΞοϓσʔτ͕৭ʑ͋Γ·ͨ͠

  5. ػցֶशؔ࿈Ξοϓσʔτ  w "84%FFQ$PNQPTFS ϓϨϏϡʔ  w $POUBDU-FOTGPS"NB[PO $POOFDU ϓϨϏϡʔ

     w "NB[PO'SBVE%FUFDUPS ϓϨϏϡʔ  w "NB[PO,FOESB ϓϨϏϡʔ  w "NB[PO$PEF(VSV ϓϨϏϡʔ  w "NB[PO"VHNFOUFE"* ϓϨϏϡʔ w"NB[PO4BHF.BLFS4UVEJP w"NB[PO4BHF.BLFS/PUFCPPLT ϓϨϏϡʔ  w"NB[PO4BHF.BLFS%FCVHHFS w"NB[PO4BHF.BLFS&YQFSJNFOUT w"NB[PO4BHF.BLFS"VUPQJMPU w"NB[PO4BHF.BLFS1SPDFTTJOH w"NB[PO4BHF.BLFS.PEFM.POJUPS w"NB[PO4BHF.BLFSͷ%FFQ(SBQI-JCSBSZର Ԡ w"NB[PO4BHF.BLFS0QFSBUPSTGPS ,VCFSOFUFT

  6. ػցֶशؔ࿈Ξοϓσʔτ  w "84%FFQ$PNQPTFS ϓϨϏϡʔ  w $POUBDU-FOTGPS"NB[PO$POOFDU ϓϨϏϡʔ 

    w "NB[PO'SBVE%FUFDUPS ϓϨϏϡʔ  w "NB[PO,FOESB ϓϨϏϡʔ  w "NB[PO$PEF(VSV ϓϨϏϡʔ  w "NB[PO"VHNFOUFE"* ϓϨϏϡʔ w"NB[PO4BHF.BLFS4UVEJP w"NB[PO4BHF.BLFS/PUFCPPLT ϓϨϏϡʔ  w"NB[PO4BHF.BLFS%FCVHHFS w"NB[PO4BHF.BLFS&YQFSJNFOUT w"NB[PO4BHF.BLFS"VUPQJMPU w"NB[PO4BHF.BLFS1SPDFTTJOH w"NB[PO4BHF.BLFS.PEFM.POJUPS w"NB[PO4BHF.BLFSͷ%FFQ(SBQI-JCSBSZ ରԠ w"NB[PO4BHF.BLFS0QFSBUPSTGPS ,VCFSOFUFT

  7. ࠓ೔࿩͢಺༰  w"NB[PO4BHF.BLFS4UVEJP w"NB[PO4BHF.BLFS/PUFCPPLT w"NB[PO4BHF.BLFS%FCVHHFS w"NB[PO4BHF.BLFS &YQFSJNFOUT w"NB[PO4BHF.BLFS"VUPQJMPU w"NB[PO4BHF.BLFS1SPDFTTJOH w"NB[PO4BHF.BLFS.PEFM.POJUPS

    w"NB[PO4BHF.BLFSͷ%FFQ(SBQI-JCSBSZରԠ w"NB[PO4BHF.BLFS0QFSBUPSTGPS,VCFSOFUFT wػցֶशʹ͓͚ΔϫʔΫϑϩʔ w"NB[PO4BHF.BLFS֓ཁ w"NB[PO4BHF.BLFSͷ֤छΞοϓσʔτ಺༰

  8.  ػցֶशͷϫʔΫϑϩʔ

  9. ػցֶशͷϫʔΫϑϩʔ 

  10. ػցֶशͷϫʔΫϑϩʔ  ࣮ݧΛ܁Γฦ͢

  11. ػցֶशͷϫʔΫϑϩʔ  ͻͨ͢Β܁Γฦ͢

  12.  "NB[PO4BHF.BLFS

  13. ͜Ε·Ͱͷ "NB[PO4BHF.BLFS  Πϯϑϥ؅ཧෆཁͰ&%"͔ΒϞσϧͷӡ༻·ͰՄೳͳ ػցֶशͷϑϧϚωʔδυαʔϏε w+VQZUFSͷϗεςΟϯά wػցֶशϞσϧͷֶशͱϋΠύʔύϥϝʔλνϡʔχϯά wػցֶशϞσϧͷϗεςΟϯάͱόονਪ࿦ w֤छ૊ΈࠐΈΞϧΰϦζϜͷ༻ҙ֤छϑϨʔϜϫʔΫʹ΋ରԠ w"NB[PO4BHF.BLFS(SPVOE5SVUIϥϕϧ͚ͮ

    Ξϊςʔγϣϯ δϣϒͷ࡞੒ͱ؅ཧ w"NB[PO4BHF.BLFS/FPϞσϧͷίϯύΠϧ w"NB[PO4BHF.BLFS4FBSDIֶशδϣϒͷݕࡧͱൺֱ

  14.  "NB[PO4BHF.BLFS ֤छΞοϓσʔτ಺༰

  15.  "NB[PO4BHF.BLFS4UVEJP

  16. "NB[PO4BHF.BLFS4UVEJP  ػցֶशͷͨΊͷ౷߹։ൃ؀ڥ *%&  w+VQZUFS-BCϕʔε w"NB[PO4BHF.BLFSͷ֤ػೳͱ࿈ܞ wίʔυϕʔεͰͷ&%"͔Β࣮ݧɺӡ༻·ͰՄೳ w440ͰͷϢʔβ؅ཧɾΞΫηε͕Մೳ wΦϋΠΦϦʔδϣϯͰར༻Մೳ

  17. "NB[PO4BHF.BLFS4UVEJP  https://aws.amazon.com/jp/blogs/news/amazon-sagemaker-studio-the-first-fully-integrated-development-environment-for-machine-learning/

  18.  "NB[PO4BHF.BLFS/PUFCPPLT ϓϨϏϡʔ

  19. "NB[PO4BHF.BLFS/PUFCPPLT ϓϨϏϡʔ  +VQZUFSϊʔτϒοΫͷར༻ͱڞ༗͕؆୯ʹ w+VQZUFS+VQZUFS-BC͕਺ඵͰ࢖͑Δ wΠϯϑϥ؅ཧ͸ෆཁ wϊʔτϒοΫͷڞ༗͕؆୯ʹͰ͖Δ wܭࢉϦιʔεͷ੾Γସ͑΋؆୯

  20. ϊʔτϒοΫͷڞ༗  https://aws.amazon.com/jp/blogs/news/amazon-sagemaker-studio-the-first-fully-integrated-development-environment-for-machine-learning/

  21.  "NB[PO4BHF.BLFS &YQFSJNFOUT

  22. "NB[PO4BHF.BLFS&YQFSJNFOUT 

  23. "NB[PO4BHF.BLFS&YQFSJNFOUT  ػցֶशδϣϒ΍࣮ݧͷ؅ཧɾൺֱΛΑΓ؆୯ʹ w࣮ݧ໊ͱτϥΠΞϧ໊Ͱ֤छδϣϒΛ؅ཧ w࣮ݧ͸τϥΠΞϧͷू߹ɺτϥΠΞϧ͸τϥΠΞϧίϯϙʔωϯτͰߏ੒ w"NB[PO4BHF.BLFS4UVEJPͱ࿈ܞ͢Δ͜ͱͰɺ(6*Ͱͷ ؅ཧͱൺֱ͕Մೳ

  24. ֶश࣌ʹ͓͚ΔϩεͷมԽΛൺֱ  https://aws.amazon.com/jp/blogs/news/amazon-sagemaker-studio-the-first-fully-integrated-development-environment-for-machine-learning/

  25.  "NB[PO4BHF.BLFS%FCVHHFS

  26. "NB[PO4BHF.BLFS%FCVHHFS 

  27. "NB[PO4BHF.BLFS%FCVHHFS  ϞσϧͷֶशΛޮ཰Խ wֶशதʹࢦఆͨ͠σʔλΛऩू wֶशঢ়ଶͷϧʔϧϕʔεͷϞχλϦϯά w$MPVE8BUDIͱ࿈ܞͨ͠ϧʔϧϕʔεͷϦΞϧλΠϜΞΫ γϣϯ w5FOTPS'MPXɺ1Z5PSDIɺ.9/FUɺ9(#PPTUʹαϙʔτ

  28. ؂ࢹϧʔϧͷఆٛɺϧʔϧҧ൓ݕग़ɺϩεΧʔϒ  https://aws.amazon.com/jp/blogs/news/amazon-sagemaker-studio-the-first-fully-integrated-development-environment-for-machine-learning/

  29.  "NB[PO4BHF.BLFS 1SPDFTTJOH

  30. "NB[PO4BHF.BLFS1SPDFTTJOH 

  31. "NB[PO4BHF.BLFS1SPDFTTJOH  લޙॲཧͱධՁॲཧͷϚωʔδυαʔϏε wεΫϦϓτͱ࣮ߦ؀ڥɺೖग़ྗ৘ใ͚ͩͰ࣮ߦՄೳ w"NB[PO4BHF.BLFSʹֶशҎ֎ͷॲཧΛΦϑϩʔυͰ͖Δ wಠࣗίϯςφͷར༻΋Մೳ

  32. εΫϦϓτྫ  https://aws.amazon.com/jp/blogs/news/amazon-sagemaker-processin-fully-managed-data-processing-and-model-evaluation/?nc1=b_rp

  33.  "NB[PO4BHF.BLFS.PEFM .POJUPS

  34. "NB[PO4BHF.BLFS.PEFM.POJUPS 

  35. "NB[PO4BHF.BLFS.PEFM.POJUPS  Ϟσϧӡ༻࣌ͷҟৗΛݕग़ wσϓϩΠͨ͠Ϟσϧͷೖग़ྗσʔλΛه࿥ɾ෼ੳ wϞχλϦϯάδϣϒͷఆظ࣮ߦ wϧʔϧϕʔεͰͷҟৗνΣοΫ w$MPVE8BUDIͱͷ࿈ܞ

  36. ͋Δಛ௃ྔͷ߹ܭ஋ͷൺֱ ج४σʔλͱϞχλϦϯάσʔλ  https://aws.amazon.com/jp/blogs/news/amazon-sagemaker-studio-the-first-fully-integrated-development-environment-for-machine-learning/

  37.  "NB[PO4BHF.BLFS"VUPQJMPU

  38. "NB[PO4BHF.BLFS"VUPQJMPU 

  39. "NB[PO4BHF.BLFS"VUPQJMPU  "VUP.- w&%"͔ΒલॲཧɺΞϧΰϦζϜͷબ୒ɺֶशɺνϡʔχϯά ·ͰࣗಈͰ࣮ߦ wߦͬͨલॲཧ΍ֶशΛ࠶ݱՄೳͳϊʔτϒοΫͷࣗಈ࡞੒

  40. "VUPQJMPUʹΑΔτϥΠΞϧҰཡ  https://aws.amazon.com/jp/blogs/news/amazon-sagemaker-studio-the-first-fully-integrated-development-environment-for-machine-learning/

  41.  "NB[PO4BHF.BLFSͷ%FFQ (SBQI-JCSBSZରԠ

  42. "NB[PO4BHF.BLFSͷ%FFQ(SBQI-JCSBSZରԠ  "NB[PO4BHF.BLFSͰάϥϑσʔλͷֶश͕ΑΓ؆ ୯ʹ w%FFQ(SBQI-JCSBSZ͸άϥϑߏ଄Λ࣋ͬͨσʔλͷͨΊͷ 044ͷਂ૚ֶशϑϨʔϜϫʔΫ wόοΫΤϯυ͸1Z5PSDIͱ.9/FUʹରԠ wίϯςφΠϝʔδ͕༻ҙ͞Ε͍ͯΔͨΊɺεΫϦϓτͱύϥ ϝʔλɺσʔλ͚ͩͰֶशՄೳ

  43. εΫϦϓτྫ  https://aws.amazon.com/jp/blogs/news/now-available-on-amazon-sagemaker-the-deep-graph-library/?nc1=b_rp

  44.  "NB[PO4BHF.BLFS0QFSBUPS GPS,VCFSOFUFT

  45. "NB[PO4BHF.BLFS0QFSBUPSGPS,VCFSOFUFT  ,VCFSOFUFTΛհͨ͠4BHF.BLFSͷར༻͕Մೳʹ wϞσϧͷֶश͚ͩͰͳ͘ɺνϡʔχϯά΍ϗεςΟϯά΋Մ ೳ w,VCFSOFUFT"1*΍LVCFDUM͔Β4BHF.BLFSͰͷδϣϒ͕࡞ ੒Մೳ

  46. ֶशδϣϒΛΧελϜϦιʔεͱͯ͠ఆٛ͠ɺ࣮ߦ  https://aws.amazon.com/jp/blogs/news/introducing-amazon-sagemaker-operators-for-kubernetes/?nc1=b_rp

  47.  ·ͱΊ

  48. "NB[PO4BHF.BLFS͕ύϫʔΞοϓʂ  %FFQ(SBQI-JCSBSZʹ΋ରԠʂ,VCFSOFUFT͔Β΋࢖͑Δʂ