[{"data":1,"prerenderedAt":133},["ShallowReactive",2],{"project-31729204":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":33,"aspects":34,"strip":36,"post_text":37,"what":38,"said":63,"maker_in_thread":80,"shape":86,"grid":88,"attempts":94,"same_week":95,"maker_shows":125,"maker_hidden":126,"reader_version":127,"card":128},"31729204","PostgresML","PostgresML integrates Hugging Face to bring SOTA models into the db","title","Show HN: PostgresML integrates Hugging Face to bring SOTA models into the db","https:\u002F\u002Fpostgresml.org\u002Fuser_guides\u002Ftransformers\u002Fpre_trained_models\u002F","postgresml.org","levkk",1655143941,2022,14,3,1755,2570,69,"shown","data","machine learning in databases","library",[24,25,26,27,28],"postgres","hugging face","transformers","in-database inference","sql machine learning","quiet",null,{"positive":32,"neutral":33,"mixed":32,"critical":32},0,2,{"performance":35},{"positive":32,"critical":32},[15,32,15],"Hello folks,\n\nIt's been a few weeks, so we thought it might be time for another update and this one is very exciting indeed.\n\nWe've added automatic integration with Hugging Face transformers! By running just a single SQL command, anyone is now able to deploy any of the state-of-the-art models into their Postgres DB for real time inference & tuning.\n\nThe list of models is long, but here it is anyway: translation, sentiment analysis, summarization, question answering and text generation.\n\nLet us know what you think!\n\nMontana & Lev",{"problem":39,"audience":30,"pricing":30,"motivation":30,"time_to_build":30,"stack":45,"openness":62,"stage":62,"team_size":62},{"value":40,"source":41,"evidence":42},"Running state-of-the-art ML models on data stored in Postgres without separate infrastructure","maker",[43],{"id":44},"post",[46,50,54],{"value":47,"source":41,"evidence":48},"PostgreSQL",[49],{"id":44},{"value":51,"source":41,"evidence":52},"Hugging Face transformers",[53],{"id":44},{"value":55,"source":56,"evidence":57},"CUDA\u002FcuBLAS for GPU acceleration","commenter",[58],{"id":59,"by":60,"text":61},"31729569","montanalow","Yep, they will use the specialized hardware by default if CUDA\u002FcuBLAS has been installed on the system.","unknown",{"praise":64,"objections":72,"requests":73,"questions":74,"alternatives":79},[65],{"value":66,"source":56,"evidence":67},"Makes Hugging Face models easy to access from Postgres",[68],{"id":69,"by":70,"text":71},"31729485","hn_nal","I’m a bit out of the ML loop, but I had no idea hugging face has made it so easy to get access to all these models. Really interesting that this is available in Postgres now!…",[],[],[75],{"value":76,"source":56,"evidence":77},"Does it work with GPUs?",[78],{"id":69,"by":70,"text":71},[],{"comments":81,"answered":81,"key":82},1,[83],{"id":84,"text":85},"31729517","Yup!",{"commenters":15,"max_depth":33,"maker_replies":81,"first_comment_mins":87,"lifespan_hours":32},18,[89,91,93],[69,32,15,90,70,32],"project\u002Fneutral\u002Fperformance",[84,32,32,92,11,81],"maker\u002Fneutral\u002Fperformance",[59,32,15,90,60,32],[],[96,101,106,111,116,121],{"id":97,"name":98,"points":99,"category":100},"31710603","Vytal",313,"security_privacy",{"id":102,"name":103,"points":104,"category":105},"31755025","anonfriendly",306,"web_saas",{"id":107,"name":108,"points":109,"category":110},"31679811","Ory Kratos",290,"infrastructure",{"id":112,"name":113,"points":114,"category":115},"31726472","Tuc",241,"dev_tools",{"id":117,"name":118,"points":119,"category":120},"31724613","Animation Magic",238,"learning_science",{"id":122,"name":123,"points":124,"category":105},"31741018","My Side Project Rocks",220,5,false,"v2",{"id":129,"name":5,"line":6,"ai":126,"cat":20,"year":13,"points":14,"comments":15,"rec":29,"s":30,"top":18,"kw":130,"maker":11,"domain":10,"tier":19,"attempts":32,"strip":131,"url":9,"mix":132,"praise":66,"objection":-1,"sub":21},31729204,[24,25,26,27,28],"303",[32,33,32,32],1793991047147]