forked from elbasir/DeepCrystal
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathTrain.py
208 lines (150 loc) · 6.75 KB
/
Train.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
import numpy as np
np.random.seed(1337)
from keras.models import load_model
import pandas as pd
import io
from keras.preprocessing.sequence import pad_sequences
from keras.preprocessing import sequence
from keras.models import Model
from keras.layers import Dense,Dropout, Input, Convolution1D, MaxPooling1D, Flatten
from keras.layers.embeddings import Embedding
from keras.layers.merge import Concatenate
from keras.optimizers import Adam
from keras.layers import merge
from keras.regularizers import l2
from keras.layers import AveragePooling1D
from sklearn.utils import shuffle
from sklearn.utils import class_weight
from keras.callbacks import ModelCheckpoint
from sklearn import metrics
from keras.models import model_from_json
from Bio import SeqIO
import argparse
X_train = []
y_train = []
parser = argparse.ArgumentParser()
parser.add_argument('file', type=argparse.FileType('r'))
args = parser.parse_args()
with args.file as file:
fasta_sequences = SeqIO.parse(file, 'fasta')
for fasta in fasta_sequences:
name, sequence = fasta.id, fasta.seq.tostring()
X_train.append(sequence)
if('diffraction' in name or 'Crysallizable' in name):
y_train.append(1)
else:
y_train.append(0)
for i in range(0, len(X_train)):
train = []
st = str(X_train[i])
trainStr =''
for ch in st:
if(ch =='B' or ch =='J' or ch =='O' or ch =='U' or ch =='Z'):
trainStr += 'X'
else:
trainStr += ch
train.append(trainStr)
X_train[i] = train
amino_acids ='ACDEFGHIKLMNPQRSTVWXY'
for i in range(0, len(X_train)):
train = []
st = str(X_train[i])
trainStr =''
for ch in st:
if(ch in amino_acids):
trainStr += ch
train.append(trainStr)
X_train[i] = train
X = []
Y = []
for i in range(len(X_train)):
if(len(X_train[i][0]) <= 800):
X.append(X_train[i][0])
Y.append(y_train[i])
X_train = np.array(X)
y_train = np.array(Y)
X_train = X_train.reshape(len(X_train),1)
embed = []
for i in range(0, len(X_train)):
length = len(X_train[i][0])
pos = []
counter = 0
st = X_train[i][0]
for c in st:
AMINO_INDEX = amino_acids.index(c)
pos.append(AMINO_INDEX)
counter += 1
while(counter < 800):
pos.append(21)
counter += 1
embed.append(pos)
embed = np.array(embed)
data,Label = shuffle(embed,y_train, random_state=2)
X_train = data
y_train = Label
class_weight = class_weight.compute_class_weight('balanced', np.unique(y_train), y_train)
class_weight_dict = dict(enumerate(class_weight))
lr = 0.001
pl = 5
l2value = 0.001
stride_ = 1
stride_max = 1
border = 'same'
lr = 0.001
pl = 5
l2value = 0.001
stride_ = 1
stride_max = 1
border = 'same'
for counter in range(1, 11):
main_input = Input(shape=(800,), dtype='int32', name='main_input')
x = Embedding(output_dim=50, input_dim=22, input_length=800)(main_input)
a = Convolution1D(64, 2, activation='relu', border_mode='same', W_regularizer=l2(l2value))(x)
apool = MaxPooling1D(pool_length=pl, stride=stride_max, border_mode='same')(a)
b = Convolution1D(64, 3, activation='relu', border_mode='same', W_regularizer=l2(l2value))(x)
bpool = MaxPooling1D(pool_length=pl, stride=stride_max, border_mode='same')(b)
c = Convolution1D(64, 8, activation='relu', border_mode='same', W_regularizer=l2(l2value))(x)
cpool = MaxPooling1D(pool_length=pl, stride=stride_max, border_mode='same')(c)
d = Convolution1D(64, 9, activation='relu', border_mode='same', W_regularizer=l2(l2value))(x)
dpool = MaxPooling1D(pool_length=pl, stride=stride_max, border_mode='same')(d)
f = Convolution1D(64, 4, activation='relu', border_mode='same', W_regularizer=l2(l2value))(x)
fpool = MaxPooling1D(pool_length=pl, stride=stride_max, border_mode='same')(f)
g = Convolution1D(64, 5, activation='relu', border_mode='same', W_regularizer=l2(l2value))(x)
gpool = MaxPooling1D(pool_length=pl, stride=stride_max, border_mode='same')(g)
h = Convolution1D(64, 6, activation='relu', border_mode='same', W_regularizer=l2(l2value))(x)
hpool = MaxPooling1D(pool_length=pl, stride=stride_max, border_mode='same')(h)
i = Convolution1D(64, 7, activation='relu', border_mode='same', W_regularizer=l2(l2value))(x)
ipool = MaxPooling1D(pool_length=pl, stride=stride_max, border_mode='same')(i)
merge2 = merge([apool, bpool, cpool, dpool,fpool,gpool,hpool, ipool], mode='concat', concat_axis=-1)
merge2 = Dropout(0.3)(merge2)
scalecnn1 = Convolution1D(64, 11, activation='relu', border_mode='same', W_regularizer=l2(l2value))(merge2)
scale1 = MaxPooling1D(pool_length=pl, stride=stride_max, border_mode='same')(scalecnn1)
scalecnn2 = Convolution1D(64, 13, activation='relu', border_mode='same', W_regularizer=l2(l2value))(merge2)
scale2 = MaxPooling1D(pool_length=pl, stride=stride_max, border_mode='same')(scalecnn2)
scalecnn3 = Convolution1D(64, 15, activation='relu', border_mode='same', W_regularizer=l2(l2value))(merge2)
scale3 = MaxPooling1D(pool_length=pl, stride=stride_max, border_mode='same')(scalecnn3)
scale = merge([scale1, scale2, scale3], mode='concat', concat_axis=-1)
scale = Dropout(0.3)(scale)
cnn1 = Convolution1D(64, 5, activation='relu', border_mode='same', W_regularizer=l2(l2value))(scale)
cnn10 = MaxPooling1D(pool_length=pl, stride=stride_max, border_mode='same')(cnn1)
cnn2 = Convolution1D(64, 9, activation='relu', border_mode='same', W_regularizer=l2(l2value))(scale)
cnn20 = MaxPooling1D(pool_length=pl, stride=stride_max, border_mode='same')(cnn2)
cnn3 = Convolution1D(64, 13, activation='relu', border_mode='same', W_regularizer=l2(l2value))(scale)
cnn30 = MaxPooling1D(pool_length=pl, stride=stride_max, border_mode='same')(cnn3)
cnn50 = merge([cnn10, cnn20, cnn30], mode='concat', concat_axis=-1)
cnn50 = Dropout(0.3)(cnn50)
x = Flatten()(cnn50)
x = Dense(256, activation='relu', name='FC', W_regularizer=l2(l2value))(x)
output = Dense(1,activation='sigmoid', name='output', W_regularizer=l2(l2value))(x)
model = Model(input=main_input, output=output)
adam = Adam(lr=lr)
model.compile(optimizer=adam, loss='binary_crossentropy', metrics=['accuracy'])
best_Weight_File="model"+str(counter)+".hdf5"
checkpoint = ModelCheckpoint(best_Weight_File, monitor='val_acc', verbose=1, save_best_only=True, mode='max')
callback_list = [checkpoint]
model.fit(X_train, y_train, validation_split=0.1, class_weight=class_weight_dict, nb_epoch=300, batch_size=64, callbacks=callback_list)
model_json = model.to_json()
with open("model"+str(counter)+".json", "w") as json_file:
json_file.write(model_json)
model.save_weights("model"+str(counter)+".h5")
print("Saved model to disk")