[{"data":1,"prerenderedAt":198},["ShallowReactive",2],{"project-15629272":3},{"id":4,"name":5,"line":6,"line_source":7,"title":8,"url":9,"domain":10,"maker":11,"posted_at":12,"year":13,"points":14,"comments":15,"rank_in_year":16,"shown_in_year":17,"top":18,"tier":19,"category":20,"subcategory":21,"format":22,"keywords":23,"reception":29,"sentiment":30,"stance_mix":31,"scored_comments":35,"aspects":36,"strip":43,"post_text":45,"what":46,"said":66,"maker_in_thread":115,"shape":126,"grid":130,"attempts":159,"same_week":160,"maker_shows":33,"maker_hidden":190,"reader_version":191,"card":192},"15629272","GAN Playground","Play with Generative Adversarial Neural Nets In-Browser","title","Show HN: GAN Playground – Play with Generative Adversarial Neural Nets In-Browser","https:\u002F\u002Freiinakano.github.io\u002Fgan-playground","reiinakano.github.io","reiinakano",1509888445,2017,97,14,471,2242,22,"notable","ai","generative adversarial networks","web_app",[24,25,26,27,28],"generative adversarial network","gan","neural network","browser","mnist","curious",-0.314,{"positive":32,"neutral":33,"mixed":32,"critical":34},0,5,4,9,{"docs":37,"design":38,"performance":40,"execution":42},{"positive":32,"critical":32},{"positive":32,"critical":39},2,{"positive":32,"critical":41},1,{"positive":32,"critical":41},[44,41,32,41,32,44,32,44,44,44,41,41,32,32],3,null,{"problem":47,"audience":45,"pricing":45,"motivation":45,"time_to_build":45,"stack":63,"openness":64,"stage":65,"team_size":65},{"value":48,"source":49,"evidence":50},"Lets people experiment with training GANs without setup","inferred",[51,55,59],{"id":52,"by":53,"text":54},"15629757","rahimnathwani","Source and README: https:\u002F\u002Fgithub.com\u002Freiinakano\u002Fgan-playground (at the time of writing, the HN post points to the demo)",{"id":56,"by":57,"text":58},"15630765","gallerdude","Small nitpick, but I feel like it should start training by default. Took me a good minute to realize why no progress was being made.",{"id":60,"by":61,"text":62},"15630205","SonOfLilit","I gave it a few minutes and got nothing. Maybe try for something even simpler than MNIST, so we can get the great immediate feedback effect of http:\u002F\u002Fplayground.tensorflow.org,…",[],"open_source","unknown",{"praise":67,"objections":68,"requests":93,"questions":102,"alternatives":110},[],[69,75,86],{"value":70,"source":71,"evidence":72},"Training takes minutes and shows no visible progress at first","commenter",[73,74],{"id":60,"by":61,"text":62},{"id":56,"by":57,"text":58},{"value":76,"source":71,"evidence":77},"Hard to understand the UI and graphs",[78,82],{"id":79,"by":80,"text":81},"15632196","sp332","Yeah ok it's pretty misleading. And if you let it run longer the generator starts getting a little better. It took me a long time to figure out the UI.",{"id":83,"by":84,"text":85},"15632284","EvgeniyZh","well you have the example on the left of graphs.",{"value":87,"source":71,"evidence":88},"Default example is misleading about generator quality",[89,92],{"id":90,"by":84,"text":91},"15632181","what you're showing is basically perfect discriminator with awful generator, isn't it?",{"id":79,"by":80,"text":81},[94,98],{"value":95,"source":71,"evidence":96},"Start training by default",[97],{"id":56,"by":57,"text":58},{"value":99,"source":71,"evidence":100},"Use a simpler dataset than MNIST",[101],{"id":60,"by":61,"text":62},[103],{"value":104,"source":71,"evidence":105},"Did training start when the user clicked train",[106],{"id":107,"by":108,"text":109},"15630375","DiThi","Did you click \"train\"? Otherwise it's just doing inference from random data.",[111],{"value":112,"source":71,"evidence":113},"TensorFlow Playground",[114],{"id":60,"by":61,"text":62},{"comments":44,"answered":41,"key":116},[117,120,123],{"id":118,"text":119},"15630217","Default parameters on MNIST should give good results in 5 minutes give or take - depending on your hardware. Can you show what you got?",{"id":121,"text":122},"15635106","Looks about right. The default setting is as basic as you can get. No transpose convolutions, batch norm, and other tricks for better GANs.",{"id":124,"text":125},"15633178","Unlike regular classification, it's hard to quantify the performance of a GAN by just looking at the loss curves. You really have to make a subjective evaluation by looking at…",{"commenters":127,"max_depth":33,"maker_replies":44,"first_comment_mins":128,"lifespan_hours":129},8,109,24,[131,133,135,139,141,143,146,148,150,152,154,156,157,158],[52,32,44,132,53,32],"project\u002Fneutral\u002Fdocs",[56,41,41,134,57,32],"project\u002Fcritical\u002Fdesign",[136,41,32,137,138,32],"15632830","commenter\u002Fneutral\u002Fdesign","zardo",[60,39,41,140,61,32],"project\u002Fcritical\u002Fperformance",[118,39,32,142,11,41],"maker\u002Fneutral\u002Fexecution",[144,39,44,145,84,32],"15635051","project\u002Fneutral\u002Fperformance",[121,39,32,147,11,41],"project\u002Fneutral\u002Fexecution",[107,39,44,149,108,32],"project\u002Fneutral\u002Fdesign",[151,39,44,149,61,32],"15630682",[153,39,44,147,80,32],"15631886",[90,39,41,155,84,32],"project\u002Fcritical\u002Fexecution",[79,39,41,134,80,32],[83,39,32,137,84,32],[124,39,32,147,11,41],[],[161,166,170,175,180,185],{"id":162,"name":163,"points":164,"category":165},"15624206","Server.js",295,"languages",{"id":167,"name":168,"points":169,"category":165},"15644430","Javalin 1.0",271,{"id":171,"name":172,"points":173,"category":174},"15645276","Orange Forum",226,"infrastructure",{"id":176,"name":177,"points":178,"category":179},"15611605","Link.fish",201,"web_saas",{"id":181,"name":182,"points":183,"category":184},"15622140","CertStream",142,"security_privacy",{"id":186,"name":187,"points":188,"category":189},"15638201","OSX-ISO",135,"dev_tools",false,"v2",{"id":193,"name":5,"line":6,"ai":190,"cat":20,"year":13,"points":14,"comments":15,"rec":29,"s":194,"top":18,"kw":195,"maker":11,"domain":10,"tier":19,"attempts":32,"strip":196,"url":9,"mix":197,"praise":-1,"objection":70,"sub":21},15629272,-0.31,[24,25,26,27,28],"31010303331100",[32,33,32,34],1793990966171]