[{"data":1,"prerenderedAt":275},["ShallowReactive",2],{"project-17377088":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":36,"aspects":37,"strip":45,"post_text":46,"what":47,"said":92,"maker_in_thread":164,"shape":169,"grid":173,"attempts":238,"same_week":239,"maker_shows":215,"maker_hidden":268,"reader_version":269,"card":270},"17377088","Kanji RNN","I trained a neural network to write Kanji","title","Show HN: I trained a neural network to write Kanji","http:\u002F\u002Fotoro.net\u002Fkanji-rnn","otoro.net","hardmaru",1529699229,2018,166,29,231,2155,11,"notable","ai","neural network demo","web_app",[24,25,26,27,28],"kanji","recurrent neural network","sketch-rnn","handwriting","tensorflow.js","debated",0.16,{"positive":32,"neutral":33,"mixed":34,"critical":35},5,1,2,4,12,{"execution":38,"idea":40,"performance":42,"docs":43,"privacy":44},{"positive":39,"critical":39},0,{"positive":41,"critical":39},3,{"positive":39,"critical":34},{"positive":39,"critical":39},{"positive":39,"critical":34},[41,39,39,39,39,39,39,32,39,32,39,39,39,33,33,41,32,32,32,41,39,33,39,39,39,39,33,39,39],null,{"problem":48,"audience":46,"pricing":46,"motivation":63,"time_to_build":46,"stack":68,"openness":89,"stage":90,"team_size":91},{"value":49,"source":50,"evidence":51},"Generates plausible made-up kanji from partial strokes drawn by the user.","inferred",[52,56,60],{"id":53,"by":54,"text":55},"17378973","zhte415","Edit: Reading some comments here - this doesn't seem to be about 'recognising' or 'predicting' existing characters, but using a dataset of characters to create a character by…",{"id":57,"by":58,"text":59},"17378840","prewett","I don't think it's your kanji writing; I can't figure out what it's supposed to do. I'm assuming it's supposed to autocomplete my kanji, but it failed at that, for both 国, 本, and…",{"id":61,"by":11,"text":62},"17377379","This was related to some work I did a few years ago but recently had time to retrain models to make it work inside the browser in an interactive setting. The dataset used to…",{"value":64,"source":65,"evidence":66},"Retrained a earlier model to run interactively in the browser; maker also was learning Chinese characters.","maker",[67],{"id":61,"by":11,"text":62},[69,77,83],{"value":70,"source":65,"evidence":71},"TensorFlow.js",[72,73],{"id":61,"by":11,"text":62},{"id":74,"by":75,"text":76},"17381861","wei_jok","I think this demo was created recently, and the old article linked in the demo is there only for background info, as the author explained in one of the comments below.…",{"value":78,"source":65,"evidence":79},"Sketch-RNN (LSTM recurrent neural network)",[80],{"id":81,"by":11,"text":82},"17379616","Interesting I think the reason that you believe the JS code is obfuscated is because the part of the code that contains the “weights” of the neural network, which contains 4-5…",{"value":84,"source":65,"evidence":85},"KanjiVG dataset",[86],{"id":87,"by":11,"text":88},"17377724","Sure, Jason. Btw this is the dataset used for training. It is a part of a larger open source project used for educational purposes, that might help you learn:…","open_source","launched","solo",{"praise":93,"objections":107,"requests":131,"questions":143,"alternatives":159},[94],{"value":95,"source":96,"evidence":97},"Clever idea and fun to play with, creates a bizarre kind of intelligence","commenter",[98,99,103],{"id":53,"by":54,"text":55},{"id":100,"by":101,"text":102},"17377868","SerLava","This is great! It gives off the feeling of a bizarre kind of intelligence. Can you please input things like 感覚的 and 意識 and 私は部屋 and 自殺 and 何私 to freak some people out?",{"id":104,"by":105,"text":106},"17381994","sbierwagen","Neat, it's like if A Book From The Sky was implemented as a neural network. https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FA_Book_from_the_Sky",[108,120],{"value":109,"source":96,"evidence":110},"Autocomplete fails for common characters and suggestions are often random",[111,115,116],{"id":112,"by":113,"text":114},"17378782","SuperNinKenDo","Did not work well at all for me, and now I'm left wondering if it's the neural networks fault, or if I'm just very bad at writing kanji now.",{"id":57,"by":58,"text":59},{"id":117,"by":118,"text":119},"17378857","kazinator","Hit or miss for me. It doesn't want to fill in anything if I draw the enclosure of \"wind\": 風. Also fails to find some of the simplest kanji; completions for just a stroke or two…",{"value":121,"source":96,"evidence":122},"Website uses non-free JavaScript",[123,127],{"id":124,"by":125,"text":126},"17378355","vat","this website uses non-free javascript",{"id":128,"by":129,"text":130},"17378765","mappu","Even in the FLOSS world this is a somewhat extremist position - see https:\u002F\u002Fwww.gnu.org\u002Fphilosophy\u002Fjavascript-trap.en.html",[132,136],{"value":133,"source":96,"evidence":134},"Bounding box or reference frame so input position matches the character",[135],{"id":117,"by":118,"text":119},{"value":137,"source":96,"evidence":138},"Add the date to the title",[139],{"id":140,"by":141,"text":142},"17381797","peterburkimsher","Please add the date to the title: (2015)",[144,148],{"value":145,"source":96,"evidence":146},"Why example stroke shapes are not correct strokes",[147],{"id":53,"by":54,"text":55},{"value":149,"source":96,"evidence":150},"References on 2x2 character structure",[151,155],{"id":152,"by":153,"text":154},"17380059","Pamar","\"a character is basically a 2x2 grid\"... Do you have any reference for this type of classification\u002Fcomposition? N.B.: I am not implying that it is incorrect - I am a dabbler in…",{"id":156,"by":157,"text":158},"17380119","jfries","I don't think it's helpful to think of the structure as being based on a 2x2 grid. A bit down on this page (random Google hit) are ways to structure a character:…",[160],{"value":161,"source":96,"evidence":162},"A Book from the Sky",[163],{"id":104,"by":105,"text":106},{"comments":41,"answered":33,"key":165},[166,167,168],{"id":61,"text":62},{"id":87,"text":88},{"id":81,"text":82},{"commenters":170,"max_depth":32,"maker_replies":41,"first_comment_mins":171,"lifespan_hours":172},19,28,36,[174,176,178,180,184,185,187,189,193,195,199,201,203,205,207,208,210,213,214,216,219,220,223,226,227,229,231,233,236],[53,39,41,175,54,39],"project\u002Fmixed\u002Fexecution",[152,39,39,177,153,39],"commenter\u002Fneutral\u002Fnone",[179,39,39,177,54,39],"17381731",[181,39,39,182,183,39],"17382091","other\u002Fneutral\u002Fnone","jcl",[156,39,39,177,157,39],[186,39,39,177,54,39],"17380657",[188,33,39,182,141,39],"17380511",[190,34,32,191,192,39],"17377263","project\u002Fpositive\u002Fidea","sova",[61,34,39,194,11,33],"project\u002Fneutral\u002Fexecution",[196,34,32,197,198,39],"17377460","maker\u002Fpositive\u002Fnone","jason_slack",[87,34,39,200,11,33],"maker\u002Fneutral\u002Fnone",[202,34,39,182,141,39],"17380604",[204,34,39,177,198,39],"17385505",[112,41,33,206,113,39],"project\u002Fcritical\u002Fperformance",[57,41,33,206,58,39],[117,41,41,209,118,39],"project\u002Fmixed\u002Fperformance",[211,35,32,197,212,39],"17377509","mbeissinger",[104,32,32,191,105,39],[100,215,32,191,101,39],6,[140,217,41,218,141,39],7,"project\u002Fneutral\u002Fdocs",[74,217,39,182,75,39],[124,221,33,222,125,39],8,"project\u002Fcritical\u002Fprivacy",[224,221,39,177,225,39],"17378418","baylearn",[128,221,39,177,129,39],[228,221,39,182,113,39],"17378796",[230,221,39,182,129,39],"17378978",[232,221,33,222,113,39],"17378810",[234,221,39,177,235,39],"17379334","gitgud",[81,221,39,237,11,33],"maker\u002Fneutral\u002Fprivacy",[],[240,245,250,255,260,264],{"id":241,"name":242,"points":243,"category":244},"17365150","Why Do I Procrastinate?",611,"productivity",{"id":246,"name":247,"points":248,"category":249},"17373465","Material Dashboard",436,"dev_tools",{"id":251,"name":252,"points":253,"category":254},"17405722","B2blaze",359,"languages",{"id":256,"name":257,"points":258,"category":259},"17400164","mimic-bloomberg-menu-no-js",355,"learning_science",{"id":261,"name":262,"points":263,"category":254},"17398506","The Vue.js Handbook",262,{"id":265,"name":266,"points":267,"category":249},"17385051","Instagram-scraper",165,false,"v2",{"id":271,"name":5,"line":6,"ai":268,"cat":20,"year":13,"points":14,"comments":15,"rec":29,"s":30,"top":18,"kw":272,"maker":11,"domain":10,"tier":19,"attempts":39,"strip":273,"url":9,"mix":274,"praise":95,"objection":109,"sub":21},17377088,[24,25,26,27,28],"30000005050001135553010000100",[32,33,34,35],1794036567145]