-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathfunction_design.py
307 lines (263 loc) · 11.6 KB
/
function_design.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
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
import pandas as pd
import numpy as np
import plotly.express as px
import matplotlib.pyplot as plt
import seaborn as sns
import itertools
import warnings
warnings.filterwarnings('ignore')
global ship1
global ship2
# # read csv files
# result=pd.read_csv('ship1_total.csv')
# result=result.set_index('timestamp')
#
# #query condition data
# full_data=result.query("condition==1")
# empty_data=result.query("condition==0")
# fuction 1 clean dataset
def LoadData(path, kind):
global ship1
global ship2
if kind == 1:
ship1 = pd.read_csv(path)
ship1 = ship1.set_index('timestamp')
else:
ship2 = pd.read_csv(path)
ship2 = ship2.set_index('timestamp')
def clean(data):
names = data.columns
IOR = []
UpLimit1 = []
DownLimit1 = []
up = []
for i, col in enumerate(names):
s = data[col]
IQR = s.quantile(0.75) - s.quantile(0.25)
IOR.append(IQR)
UpLimit = s.quantile(0.75) + IQR * 1.5
DownLimit = s.quantile(0.75) - IQR * 1.5
UpLimit1.append(UpLimit)
DownLimit1.append(DownLimit)
df_des = data.describe()
df_des.loc['UpLimit'] = UpLimit1
df_des.loc['DownLimit'] = DownLimit1
new_data = data[(data['ship_FuelEfficiency'] < df_des['ship_FuelEfficiency']['UpLimit']) & (
data['ship_FuelEfficiency'] > df_des['ship_FuelEfficiency']['DownLimit'])]
return new_data
def new_data(data1, data2):
frame = [data1, data2]
new_data = pd.concat(frame, sort=False, ignore_index=True)
from sklearn.utils import shuffle
new_data = shuffle(new_data)
return new_data
# using random forest algorithm to selected five important features
def column_selection(data):
# X = data[['engine_power', 'Rot', 'Rotation', 'RudderAngle', 'SPEED_KNOTS', 'WindDiration', 'WindSpeed', 'draft',
# 'reWind', 'trim']]
# Y = data['ship_FuelEfficiency']
# from sklearn.ensemble import RandomForestRegressor
# model = RandomForestRegressor(n_estimators=100, oob_score=True, n_jobs=-1, random_state=50,
# max_features="auto", min_samples_leaf=100)
# model.fit(X, Y)
# importances = model.feature_importances_
# indices = np.argsort(importances)[::-1]
# feat_labels = X.columns
# a = []
# for f in range(X.shape[1]):
# temp = feat_labels[indices[f]]
# a.append(temp)
select_columns = ['draft', 'engine_power', 'SPEED_KNOTS', 'Rotation', 'WindSpeed']
return select_columns
def data_transform_X(data, select_columns):
model_data = data[select_columns]
from sklearn.preprocessing import StandardScaler # 实现数据标准化
scaler = StandardScaler()
data = scaler.fit_transform(model_data)
model_data = pd.DataFrame(data, columns=model_data.columns)
from sklearn.preprocessing import PolynomialFeatures # 将标准化数据进行多项式转化，最高次项为2次。
poly = PolynomialFeatures(2)
X_poly = poly.fit(model_data)
features = pd.DataFrame(poly.transform(model_data), columns=poly.get_feature_names(model_data.columns))
X_final_data = features
return X_final_data
def Y_value(data):
# 将模块目标参数：吨海里油耗量进行中心化处理
a = (data['ship_FuelEfficiency'] - data['ship_FuelEfficiency'].min()) / (
data['ship_FuelEfficiency'].max() - data['ship_FuelEfficiency'].min())
return a
def ship1_lasso_colums(data, Y):
names = data.columns
from sklearn.linear_model import LassoCV
model_lasso = LassoCV(alphas=[1, 0.1, 0.001, 0.0005]).fit(data, Y)
from sklearn.feature_selection import RFE
# rank all features, i.e continue the elimination until the last one
rfe = RFE(model_lasso, n_features_to_select=1)
rfe.fit(data, Y)
rank = rfe.ranking_
indices = np.argsort(rank)[::1]
feat_labels = data.columns
a = []
for f in range(data.shape[1]):
temp = feat_labels[indices[f]]
a.append(temp)
lasso_select_columns = a[0:8]
return lasso_select_columns
def lasso_colums_ship2(data, Y):
names = data.columns
from sklearn.linear_model import LassoCV
model_lasso = LassoCV(alphas=[1, 0.1, 0.001, 0.0005]).fit(data, Y)
from sklearn.feature_selection import RFE
# rank all features, i.e continue the elimination until the last one
rfe = RFE(model_lasso, n_features_to_select=1)
rfe.fit(data, Y)
rank = rfe.ranking_
indices = np.argsort(rank)[::1]
feat_labels = data.columns
a = []
for f in range(data.shape[1]):
temp = feat_labels[indices[f]]
a.append(temp)
lasso_select_columns = a[0:12]
return lasso_select_columns
def lasso_coef_full(data, Y, lasso_columns):
train_full_1 = data[lasso_columns]
names = train_full_1.columns
from sklearn.linear_model import LassoCV
model_lasso = LassoCV(alphas=[1, 0.1, 0.001, 0.0005]).fit(train_full_1, Y)
coef = pd.Series(model_lasso.coef_, index=names)
coef_full = coef.round(3)
return coef_full
def his_engine_data(data):
raw_data = data.describe()
em_power = raw_data['engine_power']['50%'].round(2)
em_rotation = raw_data['Rotation']['50%'].round(2)
return em_power, em_rotation
def ship1_opt_speed_full(data, coef_full, emp_full, emr_full):
raw_full = clean(data).describe()
from scipy.optimize import minimize_scalar
def f(x):
return coef_full['SPEED_KNOTS^2'] * x ** 2 + coef_full['SPEED_KNOTS'] * x + coef_full[
'engine_power SPEED_KNOTS'] * (emp_full) * x + coef_full['engine_power'] * (emp_full) + coef_full[
'engine_power^2'] * (emp_full ** 2) + coef_full['Rotation'] * (emr_full) + coef_full[
'SPEED_KNOTS Rotation'] * (emr_full) * x
res = minimize_scalar(f, method='brent')
opt_speed = res.x * raw_full['SPEED_KNOTS']['std'] + raw_full['SPEED_KNOTS']['mean']
return opt_speed
def lasso_coef_empty(data, Y, lasso_columns):
train_full_1 = data[lasso_columns]
names = train_full_1.columns
from sklearn.linear_model import LassoCV
model_lasso = LassoCV(alphas=[1, 0.1, 0.001, 0.0005]).fit(train_full_1, Y)
coef = pd.Series(model_lasso.coef_, index=names)
coef_empty = coef.round(3)
return coef_empty
def shop1_opt_speed_empty(data, coef_empty, emp_empty, emr_empty):
raw_empty = clean(data).describe()
from scipy.optimize import minimize_scalar
def f(x):
return coef_empty['SPEED_KNOTS^2'] * x ** 2 + coef_empty['SPEED_KNOTS'] * x + coef_empty[
'engine_power SPEED_KNOTS'] * (emp_empty) * x + coef_empty['engine_power'] * (emp_empty) + coef_empty[
'engine_power^2'] * (emp_empty ** 2) + coef_empty['Rotation'] * (emr_empty)
res = minimize_scalar(f, method='brent')
opt_speed = res.x * raw_empty['SPEED_KNOTS']['std'] + raw_empty['SPEED_KNOTS']['mean']
return opt_speed
def ship2_opt_speed(data, coef_, emp_full, emr_full):
raw_full = clean(data).describe()
from scipy.optimize import minimize_scalar
def f(x):
return coef_['SPEED_KNOTS^2'] * x ** 2 + coef_['SPEED_KNOTS'] * x + coef_['engine_power SPEED_KNOTS'] * (
emp_full) * x + coef_['engine_power'] * (emp_full) + coef_['engine_power^2'] * (emp_full ** 2) + coef_[
'Rotation'] * (emr_full) + coef_['SPEED_KNOTS Rotation'] * (emr_full) * x + coef_['Rotation^2'] * (
emr_full ** 2)
res = minimize_scalar(f, method='brent')
opt_speed = res.x * raw_full['SPEED_KNOTS']['std'] + raw_full['SPEED_KNOTS']['mean']
return opt_speed
def get_ship1_speed(data, LoadCondition):
full_data = data[data['condition'] == 1]
clean_full = clean(full_data)
empty_data = data[data['condition'] == 0]
clean_empty = clean(empty_data)
whole_data = new_data(clean_full, clean_empty)
select_columns = column_selection(whole_data)
if LoadCondition == 1:
X_full = data_transform_X(clean_full, select_columns)
Y_full = Y_value(clean_full)
full_lasso_columns = ship1_lasso_colums(X_full, Y_full)
coef_full = lasso_coef_full(X_full, Y_full, full_lasso_columns)
emp_full, emr_full = his_engine_data(X_full)
full_opt_speed = ship1_opt_speed_full(full_data, coef_full, emp_full, emr_full)
return full_opt_speed
if LoadCondition == 0:
X_empty = data_transform_X(clean_empty, select_columns)
Y_empty = Y_value(clean_empty)
lasso_columns_empty = ship1_lasso_colums(X_empty, Y_empty)
coef_empty = lasso_coef_empty(X_empty, Y_empty, lasso_columns_empty)
emp_empty, emr_empty = his_engine_data(X_empty)
empty_opt_speed = shop1_opt_speed_empty(empty_data, coef_empty, emp_empty, emr_empty)
return empty_opt_speed
def get_ship2_speed(data, LoadCondition):
global ship2
global ship2optspeed
full_data = data[data['condition'] == 1]
clean_full = clean(full_data)
empty_data = data[data['condition'] == 0]
clean_empty = clean(empty_data)
whole_data = new_data(clean_full, clean_empty)
select_columns = column_selection(whole_data)
if LoadCondition == 1:
X_full = data_transform_X(clean_full, select_columns)
Y_full = Y_value(clean_full)
lasso_columns_full = lasso_colums_ship2(X_full, Y_full)
coef_full = lasso_coef_full(X_full, Y_full, lasso_columns_full)
emp_full_2, emr_full_2 = his_engine_data(X_full)
ship2_full_opt = ship2_opt_speed(full_data, coef_full, emp_full_2, emr_full_2)
return ship2_full_opt
if LoadCondition == 0:
X_empty = data_transform_X(clean_empty, select_columns)
Y_empty = Y_value(clean_empty)
lasso_columns_empty = lasso_colums_ship2(X_empty, Y_empty)
coef_empty = lasso_coef_empty(X_empty, Y_empty, lasso_columns_empty)
emp_empty_2, emr_empty_2 = his_engine_data(X_empty)
ship2_empty_opt = ship2_opt_speed(empty_data, coef_empty, emp_empty_2, emr_empty_2)
return ship2_empty_opt
def get_speed(x1, x2, x3, x4):
if x3 == 1 and x4 == 1:
x1 = (x1 - 9573.89) / 1363.21
x2 = (x2 - 44.52) / 1.95
from scipy.optimize import minimize_scalar
def f(x):
return 0.022 * x ** 2 - 0.191 * x - 0.017 * (x1) * x + 0.181 * (x1) + 0.006 * (x1 ** 2) + 0.028 * (
x2) - 0.005 * (x2) * x
res = minimize_scalar(f, method='brent')
opt_speed = res.x * 1.12 + 10.32
return opt_speed
if x3 == 1 and x4 == 0:
x1 = (x1 - 7993) / 2579.95
x2 = (x2 - 43.88) / 4.92
from scipy.optimize import minimize_scalar
def f(x):
return 0.027 * x ** 2 - 0.167 * x - 0.042 * (x1) * x + 0.008 * (x1 ** 2) + 0.025 * (x2) + 0.087 * (x2)
res = minimize_scalar(f, method='brent')
opt_speed = res.x * 1.66 + 11.67
return opt_speed
if x3 == 2 and x4 == 1:
x1 = (x1 - 9798.23) / 2344.71
x2 = (x2 - 44.63) / 3.51
from scipy.optimize import minimize_scalar
def f(x):
return 0.024 * x ** 2 - 0.226 * x + 0.012 * (x1) * x + 0.145 * (x1) + 0.007 * (x1 ** 2) + 0.028 * (
x2) + 0.065 * (x2) - 0.015 * ((x2) ** 2)
res = minimize_scalar(f, method='brent')
opt_speed = res.x * 1.38 + 10.30
return opt_speed
if x3 == 2 and x4 == 0:
x1 = (x1 - 8638.723) / 2035.38
x2 = (x2 - 43.44) / 3.39
from scipy.optimize import minimize_scalar
def f(x):
return 0.007 * x ** 2 - 0.122 * x - 0.026 * (x1) * x + 0.166 * (x1) + 0.015 * (x1 ** 2) + 0.028 * (
x2) * x + 0.065 * (x2) - 0.015 * ((x2) ** 2)
res = minimize_scalar(f, method='brent')
opt_speed = res.x * 1.40 + 11.29
return opt_speed