def mixallRNNmodel(trainX,
trainY,
valX,
valY,
physical_R_input,
folds,
train_time=None):
if (train_time == 0):
x = LSTM(1024,
init='glorot_normal',
activation='relu',
recurrent_activation='hard_sigmoid')(physical_R_input)
# x = LSTM(512,init='glorot_normal',activation='relu',return_sequences=True)(physical_all_input)
x = Dropout(0.4)(x)
x = BatchNormalization()(x)
x = Dropout(0.1)(x)
physical_R_output = Dense(2,
init='glorot_normal',
activation='softmax',
W_regularizer=l2(0.001))(x)
mixallRNNmodel = Model(physical_R_input, physical_R_output)
optimization = 'Nadam'
mixallRNNmodel.compile(loss='binary_crossentropy',
optimizer=optimization,
metrics=[keras.metrics.binary_accuracy])
else:
mixallRNNmodel = load_model('model/' + str(folds) + '/model/' +
str(train_time - 1) + 'RNNNetwork.h5')
if (trainY is not None):
weight_checkpointer = ModelCheckpoint(
filepath='./model/' + str(folds) + '/weight/' + str(train_time) +
'RNNweight.h5',
verbose=1,
save_best_only=True,
monitor='val_loss',
mode='min',
save_weights_only=True)
loss_checkpointer = LossModelCheckpoint(
model_file_path='model/' + str(folds) + '/model/' +
str(train_time) + 'RNNNetwork.h5',
monitor_file_path='model/' + str(folds) + '/loss/' +
str(train_time) + 'RNNloss.json',
verbose=1,
save_best_only=True,
monitor='val_loss',
mode='min')
fitHistory = mixallRNNmodel.fit(
trainX,
trainY,
batch_size=4096,
nb_epoch=50,
shuffle=True,
callbacks=[early_stopping, loss_checkpointer, weight_checkpointer],
class_weight='auto',
validation_data=(valX, valY))
return mixallRNNmodel
def HydrophobicityNetwork(trainX,
trainY,
valX,
valY,
physical_H_input,
folds,
train_time=None):
if (train_time == 0):
x = core.Flatten()(physical_H_input)
x = BatchNormalization()(x)
x = Dense(1024, init='glorot_normal', activation='relu')(x)
x = BatchNormalization()(x)
x = Dropout(0.4)(x)
x = Dense(512, init='glorot_normal', activation='relu')(x)
x = BatchNormalization()(x)
x = Dropout(0.3)(x)
x = Dense(256, init='glorot_normal', activation='relu')(x)
x = BatchNormalization()(x)
x = Dropout(0.2)(x)
x = Dense(128, init='glorot_normal', activation='relu')(x)
x = BatchNormalization()(x)
x = Dropout(0.1)(x)
physical_H_output = Dense(2,
init='glorot_normal',
activation='softmax',
W_regularizer=l2(0.001))(x)
HydrophobicityNetwork = Model(physical_H_input, physical_H_output)
# optimization = SGD(lr=0.01, momentum=0.9, nesterov= True)
optimization = 'Nadam'
HydrophobicityNetwork.compile(loss='binary_crossentropy',
optimizer=optimization,
metrics=[keras.metrics.binary_accuracy])
else:
HydrophobicityNetwork = load_model('model/' + str(folds) + '/model/' +
str(train_time - 1) +
'HydrophobicityNetwork.h5')
if (trainY is not None):
weight_checkpointer = ModelCheckpoint(
filepath='./model/' + str(folds) + '/weight/' + str(train_time) +
'Hydrophobicityweight.h5',
verbose=1,
save_best_only=True,
monitor='val_loss',
mode='min',
save_weights_only=True)
early_stopping = EarlyStopping(monitor='val_loss',
mode='min',
patience=50)
loss_checkpointer = LossModelCheckpoint(
model_file_path='model/' + str(folds) + '/model/' +
str(train_time) + 'HydrophobicityNetwork.h5',
monitor_file_path='model/' + str(folds) + '/loss/' +
str(train_time) + 'Hydrophobicityloss.json',
verbose=1,
save_best_only=True,
monitor='val_loss',
mode='min')
fitHistory = HydrophobicityNetwork.fit(
trainX,
trainY,
batch_size=512,
epochs=5000,
verbose=2,
validation_data=(valX, valY),
shuffle=True,
class_weight='auto',
callbacks=[early_stopping, loss_checkpointer, weight_checkpointer])
return HydrophobicityNetwork
def OtherNetwork(trainX,
trainY,
valX,
valY,
physical_O_input,
folds,
train_time=None):
if (train_time == 0):
x = core.Flatten()(physical_O_input)
x = BatchNormalization()(x)
x = Dense(256, init='glorot_normal', activation='relu')(x)
x = BatchNormalization()(x)
x = Dropout(0.2)(x)
x = Dense(128, init='glorot_normal', activation='relu')(x)
x = BatchNormalization()(x)
x = Dropout(0.2)(x)
x = Dense(64, init='glorot_normal', activation='relu')(x)
x = BatchNormalization()(x)
x = Dropout(0.1)(x)
x = Dense(32, init='glorot_normal', activation='relu')(x)
x = BatchNormalization()(x)
physical_O_output = Dense(2,
init='glorot_normal',
activation='softmax',
W_regularizer=l2(0.001))(x)
OtherNetwork = Model(physical_O_input, physical_O_output)
optimization = Nadam(lr=0.0001,
beta_1=0.9,
beta_2=0.999,
epsilon=1e-08,
schedule_decay=0.004)
OtherNetwork.compile(loss='binary_crossentropy',
optimizer=optimization,
metrics=[keras.metrics.binary_accuracy])
else:
OtherNetwork = load_model('model/' + str(folds) + '/model/' +
str(train_time - 1) + 'OtherNetwork.h5')
if (trainY is not None):
weight_checkpointer = ModelCheckpoint(
filepath='./model/' + str(folds) + '/weight/' + str(train_time) +
'Otherweight.h5',
verbose=1,
save_best_only=True,
monitor='val_loss',
mode='min',
save_weights_only=True)
early_stopping = EarlyStopping(monitor='val_loss',
mode='min',
patience=200)
loss_checkpointer = LossModelCheckpoint(
model_file_path='model/' + str(folds) + '/model/' +
str(train_time) + 'OtherNetwork.h5',
monitor_file_path='model/' + str(folds) + '/loss/' +
str(train_time) + 'Otherloss.json',
verbose=1,
save_best_only=True,
monitor='val_loss',
mode='min')
fitHistory = OtherNetwork.fit(
trainX,
trainY,
batch_size=512,
epochs=5000,
verbose=2,
validation_data=(valX, valY),
shuffle=True,
class_weight='auto',
callbacks=[early_stopping, loss_checkpointer, weight_checkpointer])
return OtherNetwork
def OnehotNetwork(trainX,
trainY,
valX,
valY,
Oneofkey_input,
folds,
train_time=None):
if (train_time == 0):
# x = conv.Convolution1D(201, 2, init='glorot_normal', W_regularizer=l1(0), border_mode="same")(Oneofkey_input)
# x = Dropout(0.4)(x)
# x = Activation('softsign')(x)
x = conv.Convolution1D(101,
3,
init='glorot_normal',
W_regularizer=l2(0),
border_mode="same")(Oneofkey_input)
x = Dropout(0.4)(x)
x = Activation('relu')(x)
x = conv.Convolution1D(101,
5,
init='glorot_normal',
W_regularizer=l2(0),
border_mode="same")(x)
x = Dropout(0.4)(x)
x = Activation('relu')(x)
x = conv.Convolution1D(101,
7,
init='glorot_normal',
W_regularizer=l2(0),
border_mode="same")(x)
x = Dropout(0.4)(x)
x = Activation('relu')(x)
x = core.Flatten()(x)
x = BatchNormalization()(x)
# x = Dense(256, init='glorot_normal', activation='relu')(x)
# x = Dropout(0.3)(x)
x = Dense(128, init='glorot_normal', activation="relu")(x)
x = Dropout(0)(x)
Oneofkey_output = Dense(2,
init='glorot_normal',
activation='softmax',
W_regularizer=l2(0.001))(x)
OnehotNetwork = Model(Oneofkey_input, Oneofkey_output)
optimization = 'Nadam'
OnehotNetwork.compile(loss='binary_crossentropy',
optimizer=optimization,
metrics=[keras.metrics.binary_accuracy])
else:
OnehotNetwork = load_model('model/' + str(folds) + '/model/' +
str(train_time - 1) + 'OnehotNetwork.h5')
if (trainY is not None):
weight_checkpointer = ModelCheckpoint(
filepath='./model/' + str(folds) + '/weight/' + str(train_time) +
'Onehotweight.h5',
verbose=1,
save_best_only=True,
monitor='val_loss',
mode='min',
save_weights_only=True)
early_stopping = EarlyStopping(monitor='val_loss',
mode='min',
patience=50)
loss_checkpointer = LossModelCheckpoint(
model_file_path='model/' + str(folds) + '/model/' +
str(train_time) + 'OnehotNetwork.h5',
monitor_file_path='model/' + str(folds) + '/loss/' +
str(train_time) + 'Onehotloss.json',
verbose=1,
save_best_only=True,
monitor='val_loss',
mode='min')
fitHistory = OnehotNetwork.fit(
trainX,
trainY,
batch_size=512,
epochs=5000,
verbose=2,
validation_data=(valX, valY),
shuffle=True,
class_weight='auto',
callbacks=[early_stopping, loss_checkpointer, weight_checkpointer])
return OnehotNetwork
def AlphaturnpropensityNetwork(trainX,
trainY,
valX,
valY,
physical_A_input,
folds,
train_time=None):
if (train_time == 0):
x = conv.Convolution1D(201,
2,
init='glorot_normal',
W_regularizer=l2(0),
border_mode="same")(physical_A_input)
x = Dropout(0.4)(x)
x = Activation('softsign')(x)
x = conv.Convolution1D(151,
3,
init='glorot_normal',
W_regularizer=l2(0),
border_mode="same")(x)
x = Dropout(0.3)(x)
x = Activation('softsign')(x)
x = conv.Convolution1D(101,
5,
init='glorot_normal',
W_regularizer=l2(0),
border_mode="same")(x)
x = Dropout(0.2)(x)
x = Activation('softsign')(x)
x = conv.Convolution1D(51,
7,
init='glorot_normal',
W_regularizer=l2(0),
border_mode="same")(x)
x = Dropout(0.1)(x)
x = Activation('softsign')(x)
x = core.Flatten()(x)
x = BatchNormalization()(x)
physical_A_output = Dense(2,
init='glorot_normal',
activation='softmax',
W_regularizer=l2(0.001))(x)
AlphaturnpropensityNetwork = Model(physical_A_input, physical_A_output)
optimization = keras.optimizers.Nadam(lr=0.0005,
beta_1=0.9,
beta_2=0.999,
epsilon=1e-08,
schedule_decay=0.004)
AlphaturnpropensityNetwork.compile(
loss='binary_crossentropy',
optimizer=optimization,
metrics=[keras.metrics.binary_accuracy])
else:
AlphaturnpropensityNetwork = load_model(
'model/' + str(folds) + '/model/' + str(train_time - 1) +
'AlphaturnpropensityNetwork.h5')
if (trainY is not None):
weight_checkpointer = ModelCheckpoint(
filepath='./model/' + str(folds) + '/weight/' + str(train_time) +
'Alphaturnpropensityweight.h5',
verbose=1,
save_best_only=True,
monitor='val_loss',
mode='min',
save_weights_only=True)
early_stopping = EarlyStopping(monitor='val_loss',
mode='min',
patience=70)
loss_checkpointer = LossModelCheckpoint(
model_file_path='model/' + str(folds) + '/model/' +
str(train_time) + 'AlphaturnpropensityNetwork.h5',
monitor_file_path='model/' + str(folds) + '/loss/' +
str(train_time) + 'Alphaturnpropensityloss.json',
verbose=1,
save_best_only=True,
monitor='val_loss',
mode='min')
fitHistory = AlphaturnpropensityNetwork.fit(
trainX,
trainY,
batch_size=512,
epochs=5000,
verbose=2,
validation_data=(valX, valY),
shuffle=True,
class_weight='auto',
callbacks=[early_stopping, loss_checkpointer, weight_checkpointer])
return AlphaturnpropensityNetwork
def Capsnet_main(trainX,
trainY,
valX=None,
valY=None,
nb_classes=2,
nb_epoch=500,
earlystop=None,
weights=None,
compiletimes=0,
compilemodels=None,
lr=0.001,
lrdecay=1,
batch_size=500,
lam_recon=0.392,
routings=3,
modeltype=5,
class_weight=None,
activefun='linear',
power=2,
predict=False,
outputweights=None,
monitor_file=None,
save_best_only=True,
load_average_weight=False):
print(trainX.shape)
if len(trainX.shape) > 3:
trainX.shape = (trainX.shape[0], trainX.shape[2], trainX.shape[3])
if (valX is not None):
print(valX.shape)
if len(valX.shape) > 3:
valX.shape = (valX.shape[0], valX.shape[2], valX.shape[3])
if (
earlystop is not None
): #use early_stop to control nb_epoch there must contain a validation if not provided will select one
early_stopping = EarlyStopping(monitor='val_loss', patience=earlystop)
nb_epoch = 10000
lr_decay = callbacks.LearningRateScheduler(
schedule=lambda epoch: lr * (lrdecay**epoch))
if compiletimes == 0:
model = CapsNet(input_shape=trainX.shape[1:],
n_class=nb_classes,
routings=routings,
modeltype=modeltype)
if "crossentropy" in modeltype:
#model.compile(optimizer=optimizers.Adam(lr=lr,epsilon=1e-08),loss=['binary_crossentropy', 'mse'],loss_weights=[1., lam_recon],metrics={'capsnet': 'accuracy'})
model.compile(optimizer=optimizers.Adam(lr=lr, epsilon=1e-08),
loss='binary_crossentropy',
metrics=['accuracy'])
else:
#model.compile(optimizer=optimizers.Adam(lr=lr,epsilon=1e-08),loss=[margin_loss, 'mse'],loss_weights=[1., lam_recon],metrics={'capsnet': 'accuracy'})
#if Radam_flag:
# model.compile(optimizer=RAdam(),loss=margin_loss,metrics=['accuracy'])
#else:
model.compile(optimizer=optimizers.Adam(lr=lr, epsilon=1e-08),
loss=margin_loss,
metrics=['accuracy'])
else:
model = compilemodels
if (predict is False):
if (weights is not None and compiletimes == 0):
print("load weights:" + weights)
model.load_weights(weights)
print("##################save_best_only " + str(save_best_only))
weight_checkpointer = LossModelCheckpoint(
model_file_path=outputweights + '_iteration' + str(compiletimes),
monitor_file_path=monitor_file + '_iteration' + str(compiletimes) +
'.json',
verbose=1,
save_best_only=save_best_only,
monitor='val_loss',
mode='min',
save_weights_only=True)
if valX is not None:
if (earlystop is None):
history = model.fit(trainX,
trainY,
batch_size=batch_size,
epochs=nb_epoch,
validation_data=[valX, valY],
class_weight=class_weight,
callbacks=[lr_decay, weight_checkpointer])
else:
history = model.fit(
trainX,
trainY,
batch_size=batch_size,
epochs=nb_epoch,
validation_data=[valX, valY],
callbacks=[early_stopping, lr_decay, weight_checkpointer],
class_weight=class_weight)
else:
history = model.fit(trainX,
trainY,
batch_size=batch_size,
epochs=nb_epoch,
class_weight=class_weight,
callbacks=[lr_decay, weight_checkpointer])
if load_average_weight:
if save_best_only:
last_weights = model.get_weights()
model.load_weights(
outputweights + '_iteration' + str(compiletimes)
) #every iteration need to reload the best model for next run
saved_weights = model.get_weights()
avg_merged_weights = list()
for layer in range(len(last_weights)):
avg_merged_weights.append(
1 / 2 * (last_weights[layer] + saved_weights[layer]))
model.set_weights(avg_merged_weights)
else:
model.load_weights(
outputweights + '_iteration' + str(compiletimes)
) #every iteration need to reload the best model for next run
#if load_average_weight:
# if save_best_only:
# last_weights =model.get_weights()
# model.load_weights(outputweights+'_iteration'+str(compiletimes)) #every iteration need to reload the best model for next run
# saved_weights = model.get_weights()
# avg_merged_weights = 1/2* Add()([last_weights,saved_weights]) #don't knwow if work
# model.set_weights(avg_merged_weights)
#else:
# model.load_weights(outputweights+'_iteration'+str(compiletimes)) #every iteration need to reload the best model for next run
return model
def AlphaturnpropensityNetwork(
train_physicalXa,
trainY,
val_physicalXa,
valY,
#pre_train_total_path = 'model/pretrain.h5',
train_time=None,
compilemodels=None):
physical_A_input = Input(shape=(train_physicalXa.shape[1],
train_physicalXa.shape[2])) #49*118=5782
early_stopping = EarlyStopping(monitor='val_loss', mode='min', patience=10)
if (train_time == 0):
x = conv.Convolution1D(201,
2,
init='glorot_normal',
W_regularizer=l2(0),
border_mode="same")(physical_A_input)
x = Dropout(0.4)(x)
x = Activation('relu')(x)
x = conv.Convolution1D(151,
3,
init='glorot_normal',
W_regularizer=l2(0),
border_mode="same")(x)
x = Dropout(0.3)(x)
x = Activation('relu')(x)
x = conv.Convolution1D(101,
5,
init='glorot_normal',
W_regularizer=l2(0),
border_mode="same")(x)
x = Dropout(0.2)(x)
x = Activation('relu')(x)
x = conv.Convolution1D(51,
7,
init='glorot_normal',
W_regularizer=l2(0),
border_mode="same")(x)
x = Dropout(0.1)(x)
x = Activation('relu')(x)
x = core.Flatten()(x)
x = BatchNormalization()(x)
physical_A_output = Dense(2,
init='glorot_normal',
activation='softmax',
W_regularizer=l2(0.001))(x)
AlphaturnpropensityNetwork = Model(physical_A_input, physical_A_output)
optimization = 'Nadam'
AlphaturnpropensityNetwork.compile(
loss='binary_crossentropy',
optimizer=optimization,
metrics=[keras.metrics.binary_accuracy])
else:
AlphaturnpropensityNetwork = load_model(
"model/" + str(train_time - 1) +
'model/AlphaturnpropensityNetwork.h5')
if (trainY is not None):
#checkpointer = ModelCheckpoint(filepath="model/"+str(train_time)+'model/AlphaturnpropensityNetwork.h5',verbose=1,save_best_only=True,monitor='val_loss',mode='min')
weight_checkpointer = ModelCheckpoint(
filepath="model/" + str(train_time) +
'modelweight/AlphaturnpropensityNetworkweight.h5',
verbose=1,
save_best_only=True,
monitor='val_loss',
mode='min',
save_weights_only=True)
loss_checkpointer = LossModelCheckpoint(
model_file_path="model/" + str(train_time) +
'model/AlphaturnpropensityNetwork.h5',
monitor_file_path="model/loss/" + str(train_time) +
"Anetloss.json",
verbose=1,
save_best_only=True,
monitor='val_loss',
mode='min')
AfitHistory = AlphaturnpropensityNetwork.fit(
train_physicalXa,
trainY,
batch_size=4096,
nb_epoch=50,
shuffle=True,
callbacks=[early_stopping, loss_checkpointer, weight_checkpointer],
class_weight='auto',
validation_data=(val_physicalXa, valY))
AlphaturnpropensityNetwork.save("model/" + str(train_time) +
'model/1AlphaturnpropensityNetwork.h5')
return AlphaturnpropensityNetwork
def OnehotNetwork(
train_oneofkeyX,
trainY,
val_oneofkeyX,
valY,
#pre_train_total_path = 'model/pretrain.h5',
train_time=None,
compilemodels=None):
Oneofkey_input = Input(shape=(train_oneofkeyX.shape[1],
train_oneofkeyX.shape[2])) #49*21=1029
early_stopping = EarlyStopping(monitor='val_loss', mode='min', patience=10)
if (train_time == 0):
x = conv.Convolution1D(201,
2,
init='glorot_normal',
W_regularizer=l1(0),
border_mode="same")(Oneofkey_input)
x = Dropout(0.4)(x)
x = Activation('relu')(x)
x = conv.Convolution1D(151,
3,
init='glorot_normal',
W_regularizer=l2(0),
border_mode="same")(x)
x = Dropout(0.4)(x)
x = Activation('relu')(x)
x = conv.Convolution1D(151,
5,
init='glorot_normal',
W_regularizer=l2(0),
border_mode="same")(x)
x = Dropout(0.4)(x)
x = Activation('relu')(x)
x = conv.Convolution1D(101,
7,
init='glorot_normal',
W_regularizer=l2(0),
border_mode="same")(x)
x = Dropout(0.4)(x)
x = Activation('relu')(x)
x = core.Flatten()(x)
x = BatchNormalization()(x)
x = Dense(256, init='glorot_normal', activation='relu')(x)
x = Dropout(0.298224)(x)
x = Dense(128, init='glorot_normal', activation="relu")(x)
x = Dropout(0)(x)
Oneofkey_output = Dense(2,
init='glorot_normal',
activation='softmax',
W_regularizer=l2(0.001))(x)
OnehotNetwork = Model(Oneofkey_input, Oneofkey_output)
optimization = 'Nadam'
OnehotNetwork.compile(loss='binary_crossentropy',
optimizer=optimization,
metrics=[keras.metrics.binary_accuracy])
else:
OnehotNetwork = load_model("model/" + str(train_time - 1) +
'model/OnehotNetwork.h5')
if (trainY is not None):
#checkpointer = ModelCheckpoint(filepath="model/"+str(train_time)+'model/OnehotNetwork.h5',verbose=1,save_best_only=True,monitor='val_loss',mode='min')
weight_checkpointer = ModelCheckpoint(
filepath="model/" + str(train_time) +
'modelweight/OnehotNetworkweight.h5',
verbose=1,
save_best_only=True,
monitor='val_loss',
mode='min',
save_weights_only=True)
loss_checkpointer = LossModelCheckpoint(
model_file_path="model/" + str(train_time) +
'model/OnehotNetwork.h5',
monitor_file_path="model/loss/" + str(train_time) +
"onehotloss.json",
verbose=1,
save_best_only=True,
monitor='val_loss',
mode='min')
onehotfitHistory = OnehotNetwork.fit(
train_oneofkeyX,
trainY,
batch_size=4096,
nb_epoch=50,
shuffle=True,
callbacks=[early_stopping, loss_checkpointer, weight_checkpointer],
class_weight='auto',
validation_data=(val_oneofkeyX, valY))
OnehotNetwork.save("model/" + str(train_time) +
'model/1OnehotNetwork.h5')
return OnehotNetwork
