def prediction(inputfile, model):
#read in data
try:
data = pd.read_csv(inputfile, header=None)
print("Data loaded")
except:
print("Dataset could not be loaded. Is the dataset missing?")
data.columns = [
'AAGE', 'ACLSWKR', 'ADTIND', 'ADTOCC', 'AHGA', 'AHRSPAY', 'AHSCOL',
'AMARITL', 'AMJIND', 'AMJOCC', 'ARACE', 'AREORGN', 'ASEX', 'AUNMEM',
'AUNTYPE', 'AWKSTAT', 'CAPGAIN', 'CAPLOSS', 'DIVVAL', 'FILESTAT',
'GRINREG', 'GRINST', 'HHDFMX', 'HHDREL', 'MIGMTR1', 'MIGMTR3',
'MIGMTR4', 'MIGSAME', 'MIGSUN', 'NOEMP', 'PARENT', 'PEFNTVTY',
'PEMNTVTY', 'PENATVTY', 'PRCITSHP', 'SEOTR', 'VETQVA', 'VETYN',
'WKSWORK', 'YEAR'
]
# preprocess data
data = preprocess(data)
# load model file
try:
loaded_model = pickle.load(open(model, 'rb'))
print("Model loaded")
except:
print("Model could not be loaded. Is the model missing?")
# predict run
result = loaded_model.predict(data)
# show result
print(result)
def prediction(inputfile, model):
#read in data
try:
data = pd.read_csv(inputfile, header=None)
print("Data loaded")
except:
print("Dataset could not be loaded. Is the dataset missing?")
data.columns = [
'AAGE', 'ACLSWKR', 'ADTIND', 'ADTOCC', 'AHGA', 'AHRSPAY', 'AHSCOL',
'AMARITL', 'AMJIND', 'AMJOCC', 'ARACE', 'AREORGN', 'ASEX', 'AUNMEM',
'AUNTYPE', 'AWKSTAT', 'CAPGAIN', 'CAPLOSS', 'DIVVAL', 'FILESTAT',
'GRINREG', 'GRINST', 'HHDFMX', 'HHDREL', 'MARSUPWT', 'MIGMTR1',
'MIGMTR3', 'MIGMTR4', 'MIGSAME', 'MIGSUN', 'NOEMP', 'PARENT',
'PEFNTVTY', 'PEMNTVTY', 'PENATVTY', 'PRCITSHP', 'SEOTR', 'VETQVA',
'VETYN', 'WKSWORK', 'YEAR', 'TARGET'
]
# features for segementation model
filterCol = [
'AAGE', 'AHGA', 'ASEX', 'CAPGAIN', 'CAPLOSS', 'DIVVAL', 'NOEMP',
'WKSWORK'
]
# preprocess data
data = preprocess(data)
# filtered data
fdata = data[filterCol]
# predict data
pdata = fdata
# apply PCA by fitting the predict data with only two dimensions
pca = PCA(n_components=2)
pca.fit(pdata)
reduced_data = pca.transform(pdata)
reduced_data = pd.DataFrame(reduced_data,
columns=['Dimension 1', 'Dimension 2'])
# load model file
try:
loaded_model = pickle.load(open(model, 'rb'))
print("Model loaded")
except:
print("Model could not be loaded. Is the model missing?")
# predict run
result = loaded_model.predict(reduced_data)
# show result
print(result)
import importlib
import func
import numpy as np
import pandas as pd
df_101191 = pd.read_excel("data/191_BWSC101 Release Log Form.xlsx")
df_101592 = pd.read_excel("data/592_BWSC101 Release Log Form.xlsx")
df_101607 = pd.read_excel("data/607_BWSC101 Release Log Form.xlsx")
df_101191["RTN"] = df_101191.apply(func.completeRTN, axis=1)
df_101592["RTN"] = df_101592.apply(func.completeRTN, axis=1)
df_101607["RTN"] = df_101607.apply(func.completeRTN, axis=1)
func.preprocess(df_101607, "101607proc.xlsx", "101607")
print(df_101607.shape)
func.preprocess(df_101592, "101592proc.xlsx", "101592")
print(df_101592.shape)
func.preprocess(df_101191, "101191proc.xlsx", "101191")
print(df_101191.shape)
df_101 = df_101191.append(df_101592)
df_101 = df_101.append(df_101607)
df_101 = df_101[(df_101["A3A"] == 1) | (df_101["A2A"] == 1)]
print(df_101.shape)
# TCLass
tclass = pd.read_excel(
'data/TClass Phase Action Dates All RTNs mgf A 04-10-2018.xlsm',
sheetname="All")
print('Implement Spectral & Spatial Joint Network!')
else:
raise NotImplementedError
############# load dataset(indian_pines & pavia_univ...)######################
a = load()
All_data, labeled_data, rows_num, categories, r, c, FLAG = a.load_data(
flag=args.dataset)
print('Data has been loaded successfully!')
##################### Demision reduction & normlization ######################
a = preprocess(args.dr_method, args.dr_num) #PCA & ICA
Alldata_DR = a.Dim_reduction(All_data)
print('Dimension reduction successfully!')
a = product(c, FLAG)
All_data_norm = a.normlization(All_data[:, 1:-1], args.mi, args.ma) #spec
Alldata_DR = a.normlization(Alldata_DR, args.mi, args.ma) #spat
image_data3D_DR = Alldata_DR.reshape(r, c, -1)
print('Image normlization successfully!')
print(
"Dataset could not be loaded. Is the dataset missing? Please use -i inputfile"
)
data.columns = [
'AAGE', 'ACLSWKR', 'ADTIND', 'ADTOCC', 'AHGA', 'AHRSPAY', 'AHSCOL',
'AMARITL', 'AMJIND', 'AMJOCC', 'ARACE', 'AREORGN', 'ASEX', 'AUNMEM',
'AUNTYPE', 'AWKSTAT', 'CAPGAIN', 'CAPLOSS', 'DIVVAL', 'FILESTAT',
'GRINREG', 'GRINST', 'HHDFMX', 'HHDREL', 'MARSUPWT', 'MIGMTR1', 'MIGMTR3',
'MIGMTR4', 'MIGSAME', 'MIGSUN', 'NOEMP', 'PARENT', 'PEFNTVTY', 'PEMNTVTY',
'PENATVTY', 'PRCITSHP', 'SEOTR', 'VETQVA', 'VETYN', 'WKSWORK', 'YEAR',
'TARGET'
]
# preprocess data
data = preprocess(data)
# balance data by SMOTE
X, y, data = handle_imbalanced_data(data)
# drop instance weight from data
instance_weight = data.MARSUPWT
data = data.drop('MARSUPWT', axis=1)
# drop label from data
X = data.drop('TARGET', axis=1)
# feature selection
'''
X : features
y : labels
data : preprocessed data