[{"data":1,"prerenderedAt":292},["ShallowReactive",2],{"project-31932434":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":27,"sentiment":28,"stance_mix":29,"scored_comments":34,"aspects":35,"strip":41,"post_text":43,"what":44,"said":84,"maker_in_thread":174,"shape":184,"grid":187,"attempts":254,"same_week":255,"maker_shows":33,"maker_hidden":284,"reader_version":285,"card":286},"31932434","Featureform","An open-source Feature Store for ML","title","Show HN: Featureform – An open-source Feature Store for ML","https:\u002F\u002Fgithub.com\u002Ffeatureform\u002Ffeatureform","github.com","simba-k",1656597609,2022,136,27,443,2570,18,"notable","ai","feature store","library",[21,24,25,26],"machine learning","feature engineering","mlops","mixed",0.172,{"positive":30,"neutral":31,"mixed":32,"critical":33},4,6,3,1,14,{"design":36,"execution":38,"idea":39,"docs":40},{"positive":37,"critical":37},0,{"positive":37,"critical":37},{"positive":33,"critical":37},{"positive":37,"critical":37},[32,37,42,37,32,32,37,37,37,37,37,42,32,37,32,37,32,37,32,37,32,37,32,42,42,33,37],5,null,{"problem":45,"audience":52,"pricing":43,"motivation":56,"time_to_build":43,"stack":60,"openness":81,"stage":82,"team_size":83},{"value":46,"source":47,"evidence":48},"Managing, sharing and serving ML model features across a data science team","maker",[49],{"id":50,"by":11,"text":51},"31932461","Hey everyone, I'm Simba Khadder, Co-Founder & CEO of Featureform. I'm super stoked to be sharing our open-source feature store with you all. At my last company, we were building…",{"value":53,"source":47,"evidence":54},"Data science and ML teams",[55],{"id":50,"by":11,"text":51},{"value":57,"source":47,"evidence":58},"Maker's last company spent most time on feature engineering with messy notebooks and shared SQL snippets",[59],{"id":50,"by":11,"text":51},[61,67,71,77],{"value":62,"source":47,"evidence":63},"Go",[64],{"id":65,"by":11,"text":66},"31939689","The client libraries are all in Python. Much like Tensorflow, even though most of the heavy-lifting is done in a different language, the client feels like native Python. The…",{"value":68,"source":47,"evidence":69},"Python client libraries",[70],{"id":65,"by":11,"text":66},{"value":72,"source":47,"evidence":73},"Redis",[74],{"id":75,"by":11,"text":76},"31939715","Imagine you're Spotify and you have a stream of user-song-timestamp triplets per listen. You'll likely want to transform it into features such as: top genre per user in last 30…",{"value":78,"source":47,"evidence":79},"Spark",[80],{"id":75,"by":11,"text":76},"open_source","unknown","company",{"praise":85,"objections":101,"requests":117,"questions":129,"alternatives":158},[86,94],{"value":87,"source":88,"evidence":89},"Architecture sits on existing infrastructure rather than adding new infra","commenter",[90],{"id":91,"by":92,"text":93},"31940674","hadsed","Architecturally this is great. Introducing more infra in an ML stack is a huge pain in the short and long run and I really love that this doesn't do that (or at least, the new…",{"value":95,"source":88,"evidence":96},"Lighter alternative to Feast",[97],{"id":98,"by":99,"text":100},"31946027","mountainriver","Looks like a nice lighter alternative to Feast. Feature stores still add a lot of complexity though, if you don’t have real time transformations they probably aren’t needed",[102,106,110],{"value":103,"source":88,"evidence":104},"API feels complicated with many unfamiliar objects",[105],{"id":91,"by":92,"text":93},{"value":107,"source":88,"evidence":108},"Feature stores add complexity and may not be needed without real-time transformations",[109],{"id":98,"by":99,"text":100},{"value":111,"source":88,"evidence":112},"Mozilla Public License 2.0 seen as unusual",[113],{"id":114,"by":115,"text":116},"31935141","suyash","Project seems to be using a weird license, therefore I'm out.",[118,122],{"value":119,"source":88,"evidence":120},"Higher-level, easier API",[121],{"id":91,"by":92,"text":93},{"value":123,"source":88,"evidence":124},"Toy example explaining the product",[125],{"id":126,"by":127,"text":128},"31939174","mrfusion","I’m not quite getting it. Would you be willing to explain it using a toy example?",[130,137,144,151],{"value":131,"source":88,"evidence":132},"How to handle columnar transforms at inference time",[133],{"id":134,"by":135,"text":136},"31940456","whoevercares","How do you solve the issue with applying columnar transform from feature engineering step as row major transformer at inference time?",{"value":138,"source":88,"evidence":139},"Whether it works across Redis and Cassandra teams",[140],{"id":141,"by":142,"text":143},"31940054","rydog_2099","We have one team that uses Redis and another that uses Cassandra, would we be able to use Featureform across both of them? Kind of like a data mesh?",{"value":145,"source":88,"evidence":146},"Why Go rather than Python",[147],{"id":148,"by":149,"text":150},"31938780","lysecret","Quick question. I see it is mostly written in go. Can you talk about why that in contrast to python?",{"value":152,"source":88,"evidence":153},"What a feature is",[154],{"id":155,"by":156,"text":157},"31935230","cal85","As someone outside the ML world, can anyone tell me what a “feature” is?",[159,170],{"value":160,"source":88,"evidence":161},"Feast",[162,166],{"id":163,"by":164,"text":165},"31932513","planetsprite","How does this compare to Feast?",{"id":167,"by":168,"text":169},"31941726","adchia","Hey! One of the maintainers of Feast here, mainly to make some friendly corrections on Feast’s functionality (e.g. Feast does in fact manage transformations) :) Feast currently…",{"value":171,"source":88,"evidence":172},"Tecton",[173],{"id":167,"by":168,"text":169},{"comments":175,"answered":42,"key":176},9,[177,180,181],{"id":178,"text":179},"31941399","Up for a call? Would love to get your Api feedback, email me at simba@featureform.com , I’ll buy you lunch :)",{"id":50,"text":51},{"id":182,"text":183},"31935253","A feature is an input to a machine learning model. You can think of a model as a black-box function that takes features and outputs a prediction: prediction = model(features) For…",{"commenters":18,"max_depth":185,"maker_replies":175,"first_comment_mins":32,"lifespan_hours":186},2,24,[188,190,192,196,198,200,204,206,207,211,214,217,220,221,222,224,226,228,231,234,235,236,238,239,243,248,251],[91,37,32,189,92,37],"project\u002Fmixed\u002Fdesign",[178,37,37,191,11,33],"maker\u002Fneutral\u002Fdocs",[193,33,42,194,195,37],"31939548","project\u002Fpositive\u002Fnone","spartee",[50,185,37,197,11,33],"maker\u002Fneutral\u002Fidea",[134,185,32,199,135,37],"project\u002Fneutral\u002Fexecution",[201,185,32,202,203,37],"31944790","project\u002Fneutral\u002Fidea","d4rkp4ttern",[155,32,37,205,156,37],"other\u002Fneutral\u002Fnone",[182,32,37,202,11,33],[208,32,37,209,210,37],"31935461","other\u002Fneutral\u002Fidea","Asafp",[212,32,37,205,213,37],"31938293","otsaloma",[215,32,37,205,216,37],"31940423","uptownfunk",[218,30,42,219,149,37],"31934392","project\u002Fpositive\u002Fidea",[126,42,32,191,127,37],[75,42,37,202,11,33],[98,31,32,223,99,37],"project\u002Fmixed\u002Fidea",[225,31,37,202,11,33],"31946190",[141,227,32,202,142,37],7,[229,227,37,230,11,33],"31940196","project\u002Fpositive\u002Fdesign",[148,232,32,233,149,37],8,"maker\u002Fneutral\u002Fexecution",[65,232,37,199,11,33],[163,175,32,202,164,37],[237,175,37,219,11,33],"31932629",[167,175,32,223,168,37],[240,241,42,194,242,37],"31939995",10,"rydog_209",[244,245,42,246,247,37],"31934646",11,"maker\u002Fpositive\u002Fnone","misbahkhan",[114,249,33,250,115,37],12,"project\u002Fcritical\u002Fnone",[252,249,37,253,11,33],"31935207","project\u002Fneutral\u002Fnone",[],[256,261,265,270,275,280],{"id":257,"name":258,"points":259,"category":260},"31988820","pamspy",235,"security_privacy",{"id":262,"name":263,"points":264,"category":260},"32002555","human eyes only",216,{"id":266,"name":267,"points":268,"category":269},"32005678","C3",182,"languages",{"id":271,"name":272,"points":273,"category":274},"31958823","Copper",146,"dev_tools",{"id":276,"name":277,"points":278,"category":279},"32001817","SPS2022",132,"web_saas",{"id":281,"name":282,"points":283,"category":274},"31936356","UI Filler",129,false,"v2",{"id":287,"name":5,"line":6,"ai":284,"cat":20,"year":13,"points":14,"comments":15,"rec":27,"s":288,"top":18,"kw":289,"maker":11,"domain":10,"tier":19,"attempts":37,"strip":290,"url":9,"mix":291,"praise":87,"objection":103,"sub":21},31932434,0.17,[21,24,25,26],"305033000005303030303035510",[30,31,32,33],1793991048059]