I do experiments with the following Keras architecture with multiple outputs:
def create_model(conv_kernels = 32, dense_nodes = 512):
model_input=Input(shape=(img_channels, img_rows, img_cols))
x = Convolution2D(conv_kernels, (3, 3), padding ='same', kernel_initializer='he_normal')(model_input)
x = Activation('relu')(x)
x = MaxPooling2D(pool_size=(2, 2))(x)
x = Convolution2D(conv_kernels, (3, 3), kernel_initializer='he_normal')(x)
x = Activation('relu')(x)
x = MaxPooling2D(pool_size=(2, 2))(x)
x = Dropout(0.25)(x)
x = Flatten()(x)
conv_out = (Dense(dense_nodes, activation='relu', kernel_constraint=maxnorm(3)))(x)
x1 = Dense(nb_classes, activation='softmax')(conv_out)
x2 = Dense(nb_classes, activation='softmax')(conv_out)
x3 = Dense(nb_classes, activation='softmax')(conv_out)
x4 = Dense(nb_classes, activation='softmax')(conv_out)
lst = [x1, x2, x3, x4]
model = Model(inputs=model_input, outputs=lst)
sgd = SGD(lr=lrate, momentum=0.9, decay=lrate/nb_epoch, nesterov=False)
model.compile(loss='categorical_crossentropy', optimizer=sgd, metrics=['accuracy'])
return model
When I try to apply grid search this way:
model = KerasClassifier(build_fn=create_model1, epochs=nb_epoch, batch_size=batch_size, verbose=0)
param_grid = dict(conv_kernels = [16, 32, 64], dense_nodes = [128, 256, 512])
grid = GridSearchCV(estimator=model, cv=4, param_grid=param_grid, n_jobs=-1)
grid_result = grid.fit(X_train, Y_train)
I get the following error message:
ValueError: Found input variables with inconsistent numbers of samples: [9416, 4]
9416 is the number of training examples, 4 is the model output number.
What is the problem here? Grid search is unavailable for multiple outputs? If so, what is the best way to apply parameter search (apart from pure manual method)?
