x = tfg.Placeholder(name="x")
target = tfg.Placeholder(name="target")
n = tfn.Multi_Layer_Network(
input_size=input_size,
hidden_size_list=[hidden_layer1_size, hidden_layer2_size],
output_size=output_size,
input_node=x,
target_node=target,
init_mean=0.0,
init_sd=0.01,
activator=tfe.Activator.ReLU.value,
optimizer=tfe.Optimizer.Adam.value,
learning_rate=0.01,
model_params_dir=model_params_dir
)
#n.draw_and_show()
data = mnist.MNIST_Data()
forward_final_output = n.feed_forward(input_data=data.test_input, is_numba=False)
print(forward_final_output.shape)
print(tff.accuracy(forward_final_output, data.test_target))
batch_size = 1000
n.learning(max_epoch=100, data=data, batch_size=batch_size, print_period=1, is_numba=False, verbose=False)
forward_final_output = n.feed_forward(input_data=data.test_input, is_numba=False)
print(tff.accuracy(forward_final_output, data.test_target))
target = tfg.Placeholder(name="target")
n = tfn.CNN(
input_dim=input_dim,
cnn_param_list=cnn_param_list,
fc_hidden_size=fc_hidden_size,
output_size=output_size,
input_node=x,
target_node=target,
conv_initializer=tfe.Initializer.Conv_Xavier_Normal.value,
initializer=tfe.Initializer.Normal.value,
init_sd=0.01,
# initializer=tfe.Initializer.Xavier.value,
activator=tfe.Activator.ReLU.value,
optimizer=tfe.Optimizer.Adam.value,
learning_rate=0.001
)
#n.draw_and_show()
data = mnist.MNIST_Data(validation_size=5000, n_splits=12, is_onehot_target=True, cnn=True)
forward_final_output = n.feed_forward(input_data=data.test_input, is_numba=False)
print(forward_final_output.shape)
print(tff.accuracy(forward_final_output, data.test_target))
batch_size = 1000
n.learning(max_epoch=5, data=data, batch_size=batch_size, print_period=1, is_numba=False, verbose=False)
forward_final_output = n.feed_forward(input_data=data.test_input, is_numba=False)
print(tff.accuracy(forward_final_output, data.test_target))
