-
Notifications
You must be signed in to change notification settings - Fork 2
/
sample_creator_unit_auto.py
146 lines (116 loc) · 4.29 KB
/
sample_creator_unit_auto.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
'''
DL models (FNN, 1D CNN and CNN-LSTM) evaluation on N-CMAPSS
12.07.2021
Hyunho Mo
'''
## Import libraries in python
import gc
import argparse
import os
import json
import logging
import sys
import h5py
import time
import matplotlib
import numpy as np
import pandas as pd
from pandas import DataFrame
import matplotlib.pyplot as plt
from matplotlib import gridspec
import math
import random
import importlib
from scipy.stats import randint, expon, uniform
import sklearn as sk
from sklearn import svm
from sklearn.utils import shuffle
from sklearn import metrics
from sklearn import preprocessing
from sklearn import pipeline
# from sklearn.metrics import mean_squared_error
from math import sqrt
import scipy.stats as stats
from utils.data_preparation_unit import df_all_creator, df_train_creator, df_test_creator, Input_Gen
seed = 0
random.seed(0)
np.random.seed(seed)
current_dir = os.path.dirname(os.path.abspath(__file__))
data_filedir = os.path.join(current_dir, 'N-CMAPSS')
data_filepath = os.path.join(current_dir, 'N-CMAPSS', 'N-CMAPSS_DS02-006.h5')
def main():
# current_dir = os.path.dirname(os.path.abspath(__file__))
parser = argparse.ArgumentParser(description='sample creator')
parser.add_argument('-w', type=int, default=10, help='window length', required=True)
parser.add_argument('-s', type=int, default=10, help='stride of window')
parser.add_argument('--sampling', type=int, default=1, help='sub sampling of the given data. If it is 10, then this indicates that we assumes 0.1Hz of data collection')
parser.add_argument('--test', type=int, default='non', help='select train or test, if it is zero, then extract samples from the engines used for training')
args = parser.parse_args()
sequence_length = args.w
stride = args.s
sampling = args.sampling
selector = args.test
# Load data
'''
W: operative conditions (Scenario descriptors)
X_s: measured signals
X_v: virtual sensors
T(theta): engine health parameters
Y: RUL [in cycles]
A: auxiliary data
'''
df_all = df_all_creator(data_filepath, sampling)
'''
Split dataframe into Train and Test
Training units: 2, 5, 10, 16, 18, 20
Test units: 11, 14, 15
'''
# units = list(np.unique(df_A['unit']))
units_index_train = [2.0, 5.0, 10.0, 16.0, 18.0, 20.0]
units_index_test = [11.0, 14.0, 15.0]
print("units_index_train", units_index_train)
print("units_index_test", units_index_test)
# if any(int(idx) == unit_index for idx in units_index_train):
# df_train = df_train_creator(df_all, units_index_train)
# print(df_train)
# print(df_train.columns)
# print("num of inputs: ", len(df_train.columns) )
# df_test = pd.DataFrame()
#
# else :
# df_test = df_test_creator(df_all, units_index_test)
# print(df_test)
# print(df_test.columns)
# print("num of inputs: ", len(df_test.columns))
# df_train = pd.DataFrame()
df_train = df_train_creator(df_all, units_index_train)
print(df_train)
print(df_train.columns)
print("num of inputs: ", len(df_train.columns) )
df_test = df_test_creator(df_all, units_index_test)
print(df_test)
print(df_test.columns)
print("num of inputs: ", len(df_test.columns))
del df_all
gc.collect()
df_all = pd.DataFrame()
sample_dir_path = os.path.join(data_filedir, 'Samples_whole')
sample_folder = os.path.isdir(sample_dir_path)
if not sample_folder:
os.makedirs(sample_dir_path)
print("created folder : ", sample_dir_path)
cols_normalize = df_train.columns.difference(['RUL', 'unit'])
sequence_cols = df_train.columns.difference(['RUL', 'unit'])
if selector == 0:
for unit_index in units_index_train:
data_class = Input_Gen (df_train, df_test, cols_normalize, sequence_length, sequence_cols, sample_dir_path,
unit_index, sampling, stride =stride)
data_class.seq_gen()
else:
for unit_index in units_index_test:
data_class = Input_Gen (df_train, df_test, cols_normalize, sequence_length, sequence_cols, sample_dir_path,
unit_index, sampling, stride =stride)
data_class.seq_gen()
if __name__ == '__main__':
main()