-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy path17_Modeling_single_cell_line.py
243 lines (161 loc) · 7.74 KB
/
17_Modeling_single_cell_line.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
# -*- coding: utf-8 -*-
"""
Created on Thu Apr 13 19:03:24 2017
@author: Antonio
"""
from Utils import *
exec(open("01_Importazione_dati_e_moduli.py").read())
# exec(open('03_Descriptive.py').read(), globals())
# exec(open("05_hypothesis_test.py").read())
exec(open("10_PCA.py").read())
name_columns = ['Y', 'Numerosità', 'Cancer type', 'SE','SSE', 'MSE',
'Root_MSE', 'RSE','RRSE',
'MAE', 'RAE','Deviance', 'Variance', 'Modello' ]
target_variables = Y_array
###################################################################
##################### Linear Regression ###########################
###################################################################
model = skl.linear_model.LinearRegression()
result_regression = []
for y in target_variables:
for t in tipo:
# t='skin'
dati_correnti = dataset[ dataset['Cancer Type'] == t]
train_dataset = create_dataset(data = dati_correnti,
target_variable = y,
explanatory_variable = X_matrix)
n = len( train_dataset)
if n>5:
print(y, t, n, '\n\n')
regression = cross_validation(splits = n,
target_variable = y,
explanatory_variable = X_matrix,
data = train_dataset )
risultati = [y, n, t] + regression + ['reg_lin']
result_regression.append( risultati )
df_regression = pd.DataFrame(result_regression)
df_regression.columns = name_columns
df_regression
#
#np.save("results/all_cell_line/all_cell_CL_regression.npy", df_regression)
#pd.DataFrame(df_regression).to_excel("results/CSV/Risultati_regression_all_CL.csv")
# np.load("results/all_cell_CL_regression.npy")
##from pandas import ExcelWriter
#
writer = pd.ExcelWriter('results/Regressione_lineare_ALL.xlsx')
df_regression.to_excel(writer)
writer.save()
###################################################################
############### Support Verctor Machine ###########################
###################################################################
result_svm_list = []
for y in target_variables:
for t in tipo:
# t='skin'
dati_correnti = dataset[ dataset['Cancer Type'] == t]
train_dataset = create_dataset(data = dati_correnti,
target_variable = y,
explanatory_variable = X_matrix)
n = len( train_dataset)
if n>5:
print(y, t, n, '\n\n')
data = train_dataset
parameters = {'kernel':('linear', 'poly', 'rbf', 'sigmoid'),
'C':[1,3,5,7,9,11,13,15,17,19],
'gamma': [0.01,0.03,0.04,0.1,0.2,0.4,0.6]}
svr = svm.SVR()
grid = GridSearchCV(svr, parameters, n_jobs = 2)
X_train = data[ explanatory_variable]
y_train = data[ y ]
print( "Scelta dei parametri \n")
start_time = time.time()
SVM = grid.fit( X_train, y_train )
print("--- %s seconds ---" % (time.time() - start_time),"\n\n")
print( grid.best_params_ ,"\n\n")
print("Stima del modello \n")
# n = len( list_data[i])
result_svm = cross_validation(splits = n,
target_variable = y,
explanatory_variable = X_matrix,
data = train_dataset,
model = SVM)
risultati = [y, n, t] + result_svm + ['SVM']
result_svm_list.append( risultati )
# regression = cross_validation(splits = n,
# target_variable = y,
# explanatory_variable = X_matrix,
# data = train_dataset )
#
# risultati = [y, n, t] + regression + ['reg_lin']
# result_regression.append( risultati )
df_svm = pd.DataFrame(result_svm_list)
df_svm.columns = name_columns
df_svm
#
# np.save("results/all_cell_line/all_cell_CL_svm.npy", df_svm)
writer = pd.ExcelWriter('results/SVM_ALL.xlsx')
df_svm.to_excel(writer)
writer.save()
###################################################################
############### Neural Network MLP ################################
###################################################################
from sklearn.grid_search import GridSearchCV
# from sklearn import ae, mlp
import sklearn.neural_network as nn
#from sklearn.neural_network import Layer
result_mlp_list = []
for y in target_variables:
for t in tipo:
# t='skin'
dati_correnti = dataset[ dataset['Cancer Type'] == t]
train_dataset = create_dataset(data = dati_correnti,
target_variable = y,
explanatory_variable = X_matrix)
n = len( train_dataset)
print(y, t, n, '\n\n')
if n>5:
data = train_dataset
parameters = {'learning_rate': ['constant', 'adaptive'],
'hidden_layer_sizes': [[64, 32, 16, 8, 4, 2],
[48, 36, 24, 12, 4],
[24, 12, 6, 3, 1],
[10, 5, 3],
[4, 2],
[2]],
'activation' : [#'identity',
'logistic'],
'max_iter': [60000] }
nn_reg = nn.MLPRegressor()
grid = GridSearchCV(nn_reg, param_grid = parameters, n_jobs = 3)
X_train = data[ explanatory_variable]
y_train = data[ y ]
print( "Scelta dei parametri Reti Neurali MLP \n")
start_time = time.time()
NeurNet = grid.fit( X_train, y_train)
print("--- %s seconds ---" % (time.time() - start_time),"\n\n")
print( NeurNet.best_params_ ,"\n\n")
print("Stima del modello \n")
result_nn = cross_validation(splits = int(n/2),
target_variable = y,
explanatory_variable = X_matrix,
data = train_dataset,
model = NeurNet)
risultati = [y, n, t] + result_nn + ['MLP']
result_mlp_list.append( risultati )
print(( risultati))
# result_svm_list.append( result_svm )
df_nn = pd.DataFrame(result_mlp_list)
df_nn.columns = name_columns
df_nn
# np.save("results/all_cell_line/all_cell_line_nn.npy", df_nn)
#
writer = pd.ExcelWriter('results/MLP_ALL.xlsx')
df_nn.to_excel(writer)
writer.save()
# regression = cross_validation(splits = n,
# target_variable = y,
# explanatory_variable = X_matrix,
# data = train_dataset )
#
# risultati = [y, n, t] + regression + ['reg_lin']
# result_regression.append( risultati )