def creat_model(self):
input = Input(shape=(self.img_h, None, 1), name='the_input')
y_pred = densenet.crnn2(input, self.n_class)
basemodel = Model(inputs=input, outputs=y_pred)
# basemodel.summary()
labels = Input(name='the_labels', shape=[None], dtype='float32')
input_length = Input(name='input_length', shape=[1], dtype='int64')
label_length = Input(name='label_length', shape=[1], dtype='int64')
loss_out = Lambda(self.ctc_lambda_func, output_shape=(1, ),
name='ctc')(
[y_pred, labels, input_length, label_length])
model = Model(inputs=[input, labels, input_length, label_length],
outputs=loss_out)
model.compile(loss={
'ctc': lambda y_true, y_pred: y_pred
},
optimizer='adam',
metrics=['accuracy'])
# model.summary()
model.load_weights(self.model_path)
return basemodel
def get_model(img_h, nclass):
input = Input(shape=(img_h, img_w, 1), name='the_input')
y_pred = densenet.crnn2(input, nclass)
basemodel = Model(inputs=input, outputs=y_pred)
basemodel.summary()
labels = Input(name='the_labels', shape=[None], dtype='float32')
input_length = Input(name='input_length', shape=[1], dtype='int64')
label_length = Input(name='label_length', shape=[1], dtype='int64')
loss_out = Lambda(ctc_lambda_func, output_shape=(1,), name='ctc')([y_pred, labels, input_length, label_length])
model = Model(inputs=[input, labels, input_length, label_length], outputs=loss_out)
model.compile(loss={'ctc': lambda y_true, y_pred: y_pred}, optimizer='adam', metrics=['accuracy'])
# model.compile(loss={'ctc': lambda y_true, y_pred: y_pred}, optimizer='RMSprop', metrics=['accuracy'])
return basemodel, model