bosch-production-line-performance's Introduction

SMU IS712 Project

xgboost method

Using the following parameters:

Commit
- e167dc06ea92ecb8f7f964f4f9df966de4daeb03
Dataset
- train_numeric
- train_date (without processing)
With feature extraction
- sample(frac=1)
- XGBClassifier(base_score=0.005)
- clf.feature_importances_>0.0005
Training
- XGBClassifier(max_depth=20, base_score=0.005)

Achieve the following result:

[   0   12   46   89  112  121  135  213  274  385  393  396  421  441  449
  527  578  783  805  883  939  946  948  960 1014 1018 1019 1029 1034 1042
 1050 1056 1105 1137 1139 1159 1161 1163 1164 1165 1166 1167 1169 1171 1173
 1174 1187 1188 1189 1191 1194 1204 1209 1210 1221 1228 1231 1237 1247 1248
 1250 1261 1270 1272 1283 1287 1293 1298 1310 1323 1348 1352 1356 1376 1378
 1379 1381 1404 1422 1424 1439 1464 1470 1475 1476 1477 1482 1484 1491 1492
 1493 1494 1495 1497 1498 1501 1502 1511 1512 1516 1517 1519 1523 1528 1531
 1533 1538 1548 1549 1550 1553 1554 1562 1575 1591 1605 1607 1620 1660 1667
 1687 1691 1693 1695 1771 1773 1784 1825 1833 1838 1839 1843 1844 1846 1847
 1848 1851 1852 1857 1858 1860 1862 1880 1881 1882 1883 1884 1886 1887 1888
 1889 1891 1893 1894 1898 1899 1900 1907 1912 1913 1914 1925 1927 1932 1934
 1936 1940 1942 1944 1948 1954 1955 1960 1961 1962 1966 1967 1971 1974 1982
 1984 1985 1987 1988 1990 1992 1993 1994 1995 1997 1999 2004 2006 2007 2010
 2018 2020 2022 2028 2034 2039 2040 2041 2048 2063 2073 2093 2107]
fold 0, ROC AUC: 0.703
fold 1, ROC AUC: 0.685
fold 2, ROC AUC: 0.688
0.69207711894
0.220070032404

Recommend Projects