一文详解LDA主题模型 - 达观数据 - SegmentFault 思否 https://segmentfault.com/a/1190000012215533
SELECT COUNT(1) FROM myu_oriv_t; #CREATE TABLE myu_oriv_t AS SELECT * FROM (SELECT t.aid,t.label,o.* FROM myt t LEFT JOIN myu_oriv o ON t.uid=o.uid) AS t; SELECT DISTINCT(age) FROM myu; 4 2 1 5 3 0 SELECT DISTINCT(gender) FROM myu; 2 1 0 SELECT DISTINCT(education) FROM myu; 7 2 5 6 1 4 3 0 SELECT DISTINCT(consumptionAbility) FROM myu; 2 1 0 SELECT DISTINCT(LBS) FROM myu; 864 SELECT DISTINCT(house) FROM myu; NULL 1 UPDATE myu_copy SET house=0 WHERE house='NULL' ; sas@DESKTOP-R MINGW64 ~/PycharmProjects/py_win_to_unix/sci $ head myu_oriv_t.csv aid,label,uid,age,gender,marriageStatus,education,consumptionAbility,LBS,house 699,-1,78508957,1,1,10,1,1,576,0 1991,-1,3637295,3,2,13,7,0,353,0 1119,-1,19229018,4,1,13 10,1,0,94,0 2013,-1,79277120,2,1,10,2,1,48,1 692,-1,41528441,1,1,10,7,1,333,0 1119,-1,62381478,4,1,10,1,0,668,0 117,-1,36832847,4,1,11,1,1,333,0 389,-1,22023604,1,1,10,1,1,85,0 432,-1,39527165,5,1,10,7,2,809,0 sas@DESKTOP-R MINGW64 ~/PycharmProjects/py_win_to_unix/sci $ tail myu_oriv_t.csv 1021,-1,73766480,5,1,11,5,0,742,0 121,-1,36907820,1,2,13 10,7,2,612,0 543,-1,36747574,4,1,11,5,1,812,0 725,-1,58308810,2,1,13 10,2,0,458,0 1599,-1,57576857,4,1,11,6,1,486,0 411,-1,47406093,2,1,13 10,7,1,950,1 1429,-1,56815168,2,1,2 13,2,1,373,1 369,-1,6620624,2,1,10,2,2,514,0 692,-1,40417124,2,1,13,2,2,809,0 1918,-1,43067224,5,1,11,6,1,333,0 sas@DESKTOP-R MINGW64 ~/PycharmProjects/py_win_to_unix/sci $
第二步：将LAD结果的属性值二（多）值化、线性化，投入计算模型 DNN