def test_multiple_branch(self):
from dlpy.sequential import Sequential
model = Sequential(self.s, model_table='Simple_CNN')
model.add(InputLayer(3, 48, 96, scale=1 / 255, random_mutation='none'))
model.add(Conv2d(64, 7, include_bias=True, act='relu'))
model.add(Pooling(2))
model.add(Conv2d(64, 3, include_bias=True, act='relu'))
model.add(Pooling(2))
model.add(Conv2d(64, 3, include_bias=True, act='relu'))
model.add(Pooling(2))
model.add(Conv2d(64, 3, include_bias=True, act='relu'))
model.add(Pooling(2))
model.add(Dense(16))
model.add(OutputLayer(n=1, act='sigmoid'))
branch = model.to_functional_model(stop_layers=model.layers[-1])
inp1 = Input(**branch.layers[0].config) # tensor
branch1 = branch(inp1) # tensor
inp2 = Input(**branch.layers[0].config) # tensor
branch2 = branch(inp2) # tensor
inp3 = Input(**branch.layers[0].config) # tensor
branch3 = branch(inp3) # tensor
triplet = OutputLayer(n=1)(branch1 + branch2 + branch3)
triplet_model = Model(self.s,
inputs=[inp1, inp2, inp3],
outputs=triplet)
triplet_model.compile()
triplet_model.print_summary()
self.assertEqual(len(triplet_model.layers), 31)
triplet_model.share_weights({'Convo.1': ['Convo.1_2', 'Convo.1_3']})
triplet_model.compile()
def test_mix_cnn_rnn_network(self):
from dlpy.applications import ResNet50_Caffe
from dlpy import Sequential
from dlpy.blocks import Bidirectional
# the case is to test if CNN and RNN model can be connect using functional api
# the model_type is expected to be RNN in 19w47.
# CNN
model = ResNet50_Caffe(self.s)
cnn_head = model.to_functional_model(stop_layers=model.layers[-1])
# RNN
model_rnn = Sequential(conn=self.s, model_table='rnn')
model_rnn.add(Bidirectional(n=100, n_blocks=2))
model_rnn.add(OutputLayer('fixed'))
f_rnn = model_rnn.to_functional_model()
# connecting
inp = Input(**cnn_head.layers[0].config)
x = cnn_head(inp)
y = f_rnn(x)
cnn_rnn = Model(self.s, inp, y)
cnn_rnn.compile()
# check type
self.assertTrue(cnn_rnn.model_type == 'RNN')
self.assertTrue(cnn_rnn.layers[-1].name == 'fixed')
f_rnn = model_rnn.to_functional_model()
# connecting
inp = Input(**cnn_head.layers[0].config)
x = cnn_head(inp)
y = f_rnn(x)
cnn_rnn = Model(self.s, inp, y)
cnn_rnn.compile()
# it should be fixed if I create f_rnn again.
self.assertTrue(cnn_rnn.layers[-1].name == 'fixed')
inp = Input(**cnn_head.layers[0].config)
x = cnn_head(inp)
y = f_rnn(x)
cnn_rnn = Model(self.s, inp, y)
cnn_rnn.compile()
# it should be fixed if I create f_rnn again.
self.assertTrue(cnn_rnn.layers[-1].name == 'fixed_2')
def test_sequential_conversion(self):
from dlpy.sequential import Sequential
model1 = Sequential(self.s)
model1.add(InputLayer(3, 224, 224))
model1.add(Conv2d(8, 7))
model1.add(Pooling(2))
model1.add(Conv2d(8, 7))
model1.add(Pooling(2))
model1.add(Dense(16))
model1.add(OutputLayer(act='softmax', n=2))
func_model = model1.to_functional_model()
func_model.compile()
func_model.print_summary()
def test_stop_layers(self):
from dlpy.sequential import Sequential
model1 = Sequential(self.s)
model1.add(InputLayer(3, 224, 224))
model1.add(Conv2d(8, 7))
model1.add(Pooling(2))
model1.add(Conv2d(8, 7))
model1.add(Pooling(2))
model1.add(Dense(16))
model1.add(OutputLayer(act='softmax', n=2))
inputlayer = model1.layers[0]
inp = Input(**inputlayer.config)
func_model = model1.to_functional_model(
stop_layers=[model1.layers[-1]])
x = func_model(inp)
out = Keypoints(n=10)(x)
func_model_keypoints = Model(self.s, inp, out)
func_model_keypoints.compile()
func_model_keypoints.print_summary()
