[Python Charity Talks in Japan] LT: ネットワーク解析とテキスト解析で見るPEP / pycharity

Title:
Exploring PEP with network analysis and text analysis
(Lightning Talk at Python Charity Talks in Japan🐍🇯🇵)

I clustered PEPs using a citation-relationship network and created a word cloud for each cluster.

You can see the results of the clustering on the following page✨
https://komo-fr.github.io/PEPCluster/

Twitter: @komo_fr ( https://twitter.com/komo_fr )
GitHub: https://github.com/komo-fr/PEPCluster
Event Page: https://pyconjp.connpass.com/event/177586/
Video: https://www.youtube.com/watch?v=o-UBokTvQjE&feature=youtu.be
-------

Python Charity Talks in JapanのLTスライドです。
PEPの引用関係ネットワークを使ってコミュニティ抽出し、各クラスタのワードクラウドを出してみました。

Transcript

1. ωοτϫʔΫղੳͱςΩετղੳ
   ͰݟΔ1&1
   1ZUIPO$IBSJUZ5BMLTJO+BQBO
   
   5PNPLP'VSVLJ
   5XJUUFS!LPNP@GS(JU)VC!LPNPGS
   

   View Slide

2. !LPNP@GS
   8IPBN*
   !LPNP@GS
   2
   ॴଐ
   w גࣜձࣾϏʔϓϥ΢υ
   ʢ1ZUIPOͰσʔλղੳͷ͓࢓ࣄʣ
   w ౦ژେֶઌ୺Պֶٕज़ݚڀηϯλʔ
   ʢݚڀऀ͕࢖͏ͨΊͷ1ZUIPOύοέʔδͷ։ൃʣ
   1Z$PO+1
   w 1Z$PO+1ʮQBOEBTͷ4UZMJOHػೳͰڧԽ͢Δ+VQZUFS࣮ݧϨϙʔτʯ
   w 1Z$PO+1ʮ*OUFSBDUJWF/FUXPSL7JTVBMJ[BUJPOVTJOH1ZUIPO
   ɹɹɹɹɹɹɹɹɹʙ/FUXPSL9#PLFIͰ1&1ͷࢀরؔ܎ΛՄࢹԽ͢Δʯ
   5PNPLP'VSVLJ
   
   LPNP G S
   

   View Slide

3. !LPNP@GS
   ࠓ೔࿩͢͜ͱ
   w 1&1ಉ࢜ͷҾ༻ؔ܎ωοτϫʔΫΛ࢖ͬͯɺ
   ݱࡏҎ্͋Δ1&1ΛΫϥελϦϯάͯ͠ΈͨΑ
   3
   w ֤ΫϥελΛςΩετղੳͯ͠ɺ
   ͲΜͳΫϥελ͕Ͱ͖ͨͷ͔ݟͯΈͨΑ
   $PNNVOJUZ%FUFDUJPO 8PSE$MPVE
   
   

   View Slide

4. .PUJWBUJPO
   

   View Slide

5. !LPNP@GS
   .PUJWBUJPO
   w 1&1 1ZUIPO&OIBODFNFOU1SPQPTBMT
   ͸
   ॏཁͳެࣜͷٕज़υΩϡϝϯτ
   w Ͱ΋ྔ͕ଟ͍ʢݱࡏɺ1&1ʣ
   w ͲΕ͕ॏཁͳ1&1ͳͷʁ
   w ͲΜͳτϐοΫ͕͋Δͷʁ
   w ·ͣ͸ͬ͘͟Γͱશମ૾Λ೺Ѳ͍ͨ͠
   5
   

   View Slide

6. !LPNP@GS
   *EFB
   PEP
   8
   Style Guide
   for Python
   PEP
   20
   The Zen of
   Python
   PEP
   257
   Docstring
   Conventions
   w 1&1ͱ1&1ͷؒʹ͸Ҿ༻ؔ܎͕͋Δ
   PEP 8͸ɺ257ͱ20ΛҾ༻͍ͯ͠Δ
   

   View Slide

7. !LPNP@GS
   *EFB
   PEP
   8
   Style Guide
   for Python
   PEP
   20
   The Zen of
   Python
   PEP
   257
   Docstring
   Conventions
   w 1&1ͱ1&1ͷؒʹ͸Ҿ༻ؔ܎͕͋Δ ϦϯΫ͕షΒΕ͍ͯΔ
   
   w Ҿ༻ؔ܎ʹ͋Δ1&1ಉ࢜͸ɺτϐοΫ͕ؔ܎͍ͯ͠Δ
   w ωοτϫʔΫղੳͷख๏Λ༻͍ͯɺ
   1&1ΛΫϥελϦϯάͯ͠ΈΑ͏ʢίϛϡχςΟநग़ʣ
   

   View Slide

8. !LPNP@GS
   ࢖ͬͨύοέʔδ
   w #FBVUJGVMTPVQʢεΫϨΠϐϯάʣ
   w 1&1͸8FC্Ͱެ։͞Ε͍ͯΔͷͰɺBλάΛநग़͢Δ
   w QBOEBTʢσʔλ੔ܗʣ
   w /FUXPSL9 άϥϑσʔλԽʣ
   w QZUIPOMPVWBJOʢΫϥελϦϯά ίϛϡχςΟநग़
   ʣ
   w XPSEDMPVEʢϫʔυΫϥ΢υʣ
   w ֤Ϋϥελͷ1&1ͷຊจʹద༻
   8
   

   View Slide

9. !LPNP@GS
   1&1ωοτϫʔΫͷશମ૾
   ৭1&1ͷεςʔλε "DDFQUFE 3FKFDUͳͲ
   
   αΠζඃҾ༻਺
   

   View Slide

10. !LPNP@GS
    ˓ͷݸݸ͕ͭͷ1&1
    ৭1&1ͷεςʔλε "DDFQUFE 3FKFDUͳͲ
    
    αΠζඃҾ༻਺
    
    
    
    1&1
    1&1
    1&1
    
    1&1
    

    View Slide

11. !LPNP@GS
    QZUIPOMPVWBJOͰΫϥελϦϯά
    ৭1&1ͷεςʔλε "DDFQUFE 3FKFDUͳͲ
    
    αΠζඃҾ༻਺
    

    View Slide

12. !LPNP@GS
    ΫϥελϦϯάͷ݁Ռ
    12
    and others…
    େ͖ͳΫϥελ ݸҎ্ͷ1&1
    
    Ϋϥελ
    খ͞ͳΫϥελ ݸͷ1&1
    
    Ϋϥελ
    ݽཱͨ͠Ϋϥελ
    ʢଞͷ1&1ͱҾ༻ؔ܎͕ͳ͍1&1ʣ
    Ϋϥελ
    

    View Slide

13. !LPNP@GS
    "Ϋϥελ1SPDFTT%PDVNFOUT
    13
    • PEP 8000-series
    • PEP 1:

    PEP Purpose Guidelines
    • PEP 12:

    Sample reStructuredText PEP
    Template
    • PEP 257:

    Docstring Conventions

    25 PEPs
    

    View Slide

14. !LPNP@GS
    "Ϋϥελ5ZQF
    14
    • PEP 8
    • PEP 482, 483, 484ଞ:

    ܕώϯτؔ܎
    • PEP 3107:

    ؔ਺Ξϊςʔγϣϯ
    • PEP 557:

    σʔλΫϥε
    42 PEPs
    

    View Slide

15. !LPNP@GS
    "Ϋϥελ1BDLBHF
    15
    36 PEPs
    • PEP 241 / 314 / 345 / 426 / 566: 

    Metadata for Python Software
    Packages / v1.1 / 1.2 / 2.0 / 2.1
    • PEP 427 / 491:

    The Wheel Binary Package 

    Format 1.0 / 1.9
    • PEP 440:

    Version Identification and
    Dependency Specification
    

    View Slide

16. !LPNP@GS
    ॴײ
    w ͦΕͳΓʹೲಘ͕͍͘෦෼΋͋Δ
    w ͍·͍ͪඍົͳͱ͜Ζ΋͋Δ
    w ϦϦʔεͷ͓஌Βͤܥͷ1&1͕ɺ
    Ͳ͔͜ͷΫϥελʹద౰ʹ೧͡ࠐ·ΕΔ
    w ଞͷ1&1ͱҾ༻ؔ܎͕ͳ͍1&1͸ʢ౰વʣΫϥελϦϯάͰ΋ݽཱ͢
    Δ
    w 1&1͕͚ͭͩͷখ͞ͳΫϥελΛ࡞͍ͬͯΔ͕ɺ
    ଞͷ6OJDPEFؔ܎ͷ1&1ͱಉ͡ΫϥελʹೖΔํ͕ࣗવͳͷͰ͸
    w ࠓޙ
    w Ҿ༻ؔ܎Ͱ͸ͳ͘
    ຊจͷ಺༰ͰΫϥελϦϯάͯ͠ൺֱ͢Δͱ໘ന͍͔΋͠Εͳ͍
    16
    

    View Slide

17. !LPNP@GS
    ࠓ೔ͷ݁Ռ
    17
    IUUQTHJUIVCDPNLPNPGS1&1$MVTUFS
    

    View Slide