tmpI, tmpJ = 0, 0
dr = DataReader()
baseDf, ansY = dr.cvtPathListToDfList(_basePath + "010_blenderXgboost_train.csv", "train")
tmpOutPath = _basePath + "010_train_last_blender.csv"
tmpFeatureBlendedAns = pd.DataFrame()
baseDf = pd.DataFrame()
tmpDfList = []
for tmpClfName in clfNameList:
dr = DataReader()
tmpPath = _basePath + "010_" + "blender" + tmpClfName + "_train.csv"
newX, tmpY = dr.cvtPathListToDfList(tmpPath, "train")
tmpDfList.append(newX)
b1 = Blender(clfNameList, tmpDfList, ansY)
b1.autoFlow(2000, tmpOutPath)
# test
finalWeight = b1._bestParamList
tmpOutPath = _basePath + "010_test_last_blender.csv"
baseDf = pd.DataFrame()
tmpI = 0
finalTestDf = pd.DataFrame()
for tmpClfName in clfNameList:
dr = DataReader()
tmpPath = _basePath + "010_" + "blender" + tmpClfName + "_test.csv"
newX, tmpY = dr.cvtPathListToDfList(tmpPath, "train")
tmpWeight = finalWeight[tmpI]
newX.multiply(tmpWeight)
if tmpI == 0:
tmpPath = _basePath + "008_submission_1_train_Random_Forest.csv"
newX, newY = dr.cvtPathListToDfList(tmpPath, "train")
tmpDfList.append(newX)
tmpPath = _basePath + "008_submission_1_train_Xgboost.csv"
newX, newY = dr.cvtPathListToDfList(tmpPath, "train")
tmpDfList.append(newX)
clfNameList = []
clfNameList.append("Extra_Trees")
clfNameList.append("K_NN")
clfNameList.append("RandomForest")
clfNameList.append("Xgboost")
b1 = Blender(clfNameList, tmpDfList, newY)
b1.autoFlow(1000, outputPath)
if doTestFlag == True:
dr.readInCSV(testPath , "test")
newX = dr._testDataFrame
#newX = pd.DataFrame(newX[newX.columns[0]])
#print newX
# 3. get all best model from newX
# fab = ModelFactory()
# fab._gridSearchFlag = True
# fab._n_iter_search = 100
# fab._expInfo = expInfo
# fab.getXgboostClf(newX, newY)
tmpPath = _basePath + "008_submission_1_train_Random_Forest.csv"
newX, newY = dr.cvtPathListToDfList(tmpPath, "train")
tmpDfList.append(newX)
tmpPath = _basePath + "008_submission_1_train_Xgboost.csv"
newX, newY = dr.cvtPathListToDfList(tmpPath, "train")
tmpDfList.append(newX)
clfNameList = []
clfNameList.append("Extra_Trees")
clfNameList.append("K_NN")
clfNameList.append("RandomForest")
clfNameList.append("Xgboost")
b1 = Blender(clfNameList, tmpDfList, newY)
b1.autoFlow(1000, outputPath)
if doTestFlag == True:
dr.readInCSV(testPath, "test")
newX = dr._testDataFrame
#newX = pd.DataFrame(newX[newX.columns[0]])
#print newX
# 3. get all best model from newX
# fab = ModelFactory()
# fab._gridSearchFlag = True
# fab._n_iter_search = 100
# fab._expInfo = expInfo
# fab.getXgboostClf(newX, newY)
tmpI, tmpJ = 0, 0
dr = DataReader()
baseDf, ansY = dr.cvtPathListToDfList(
_basePath + "010_blenderXgboost_train.csv", "train")
tmpOutPath = _basePath + "010_train_last_blender.csv"
tmpFeatureBlendedAns = pd.DataFrame()
baseDf = pd.DataFrame()
tmpDfList = []
for tmpClfName in clfNameList:
dr = DataReader()
tmpPath = _basePath + "010_" + "blender" + tmpClfName + "_train.csv"
newX, tmpY = dr.cvtPathListToDfList(tmpPath, "train")
tmpDfList.append(newX)
b1 = Blender(clfNameList, tmpDfList, ansY)
b1.autoFlow(2000, tmpOutPath)
# test
finalWeight = b1._bestParamList
tmpOutPath = _basePath + "010_test_last_blender.csv"
baseDf = pd.DataFrame()
tmpI = 0
finalTestDf = pd.DataFrame()
for tmpClfName in clfNameList:
dr = DataReader()
tmpPath = _basePath + "010_" + "blender" + tmpClfName + "_test.csv"
newX, tmpY = dr.cvtPathListToDfList(tmpPath, "train")
tmpWeight = finalWeight[tmpI]
newX.multiply(tmpWeight)
if tmpI == 0: