第二步：将LAD结果的属性值二（多）值化，投入计算模型

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