def randomSample(tupleData=None, testSize=0.2):
print('=======================================================')
print('=> Randoming data...')
if (tupleData):
X_train_names, X_test_names, y_train_name, y_test_name = \
train_test_split(np.asarray(tupleData[0]['X']), tupleData[0]['y'], test_size=testSize)
X_train_address, X_test_address, y_train_address, y_test_address = \
train_test_split(np.asarray(tupleData[1]['X']), tupleData[1]['y'], test_size=testSize)
X_train_phone, X_test_phone, y_train_phone, y_test_phone = \
train_test_split(np.asarray(tupleData[2]['X']), tupleData[2]['y'], test_size=testSize)
else:
datatuple = store.loadFeatureCSV()
X_train_names, X_test_names, y_train_name, y_test_name = \
train_test_split(np.asarray(datatuple[0][1]), datatuple[0][0], test_size=testSize)
X_train_address, X_test_address, y_train_address, y_test_address = \
train_test_split(np.asarray(datatuple[1][1]), datatuple[1][0], test_size=testSize)
X_train_phone, X_test_phone, y_train_phone, y_test_phone = \
train_test_split(np.asarray(datatuple[2][1]), datatuple[2][0], test_size=testSize)
X_train = np.append(np.append(X_train_names.tolist(),
X_train_address.tolist(),
axis=0),
X_train_phone,
axis=0)
y_train = y_train_name + y_train_address + y_train_phone
X_test = np.append(np.append(X_test_names.tolist(),
X_test_address.tolist(),
axis=0),
X_test_phone,
axis=0)
y_test = y_test_name + y_test_address + y_test_phone
print('=> Randomed data.')
return (X_train, y_train, X_test, y_test)
def randomSample(tupleData=None, testSize=0.2):
print('=======================================================')
print('=> Randoming data...')
if (tupleData):
X_train_names, X_test_names, y_train_name, y_test_name = \
train_test_split(np.asarray(tupleData[0]['X']), tupleData[0]['y'], test_size=testSize)
X_train_address, X_test_address, y_train_address, y_test_address = \
train_test_split(np.asarray(tupleData[1]['X']), tupleData[1]['y'], test_size=testSize)
X_train_phone, X_test_phone, y_train_phone, y_test_phone = \
train_test_split(np.asarray(tupleData[2]['X']), tupleData[2]['y'], test_size=testSize)
else:
datatuple = store.loadFeatureCSV()
X_train_names, X_test_names, y_train_name, y_test_name = \
train_test_split(np.asarray(datatuple[0][1]), datatuple[0][0], test_size=testSize)
X_train_address, X_test_address, y_train_address, y_test_address = \
train_test_split(np.asarray(datatuple[1][1]), datatuple[1][0], test_size=testSize)
X_train_phone, X_test_phone, y_train_phone, y_test_phone = \
train_test_split(np.asarray(datatuple[2][1]), datatuple[2][0], test_size=testSize)
X_train = np.append(np.append(X_train_names.tolist(),X_train_address.tolist(), axis=0), X_train_phone, axis=0)
y_train = y_train_name + y_train_address + y_train_phone
X_test = np.append(np.append(X_test_names.tolist(), X_test_address.tolist(), axis=0), X_test_phone, axis=0)
y_test = y_test_name + y_test_address + y_test_phone
print('=> Randomed data.')
return (X_train, y_train, X_test, y_test)
