def test_sigmoid_2_backward(self):
print('\n==================================')
print(' Test sigmoid backward ')
print('==================================')
np.random.seed(123)
x = np.random.randn(5, 5)
sigmoid = Sigmoid()
out = sigmoid.forward(x)
dx = sigmoid.backward(x, 0.0)
correct_dx = [
[-0.2048748, 0.19633044, 0.06934706, -0.22376078, -0.13318844],
[0.22297909, -0.18096644, -0.10244393, 0.21720183, -0.18059114],
[-0.15157559, -0.02362423, 0.2236528, -0.14447372, -0.10570044],
[-0.10362309, 0.1971545, 0.1984592, 0.19703887, 0.09303448],
[0.16139223, 0.22364718, -0.18937883, 0.21188431, -0.21658554]
]
e = rel_error(correct_dx, dx)
print('Relative difference:', e)
self.assertTrue(e <= 1e-7)
def test_sigmoid_1_forward(self):
print('\n==================================')
print(' Test sigmoid forward ')
print('==================================')
x = np.linspace(-1.5, 1.5, num=30).reshape(6, 5)
sigmoid = Sigmoid()
out = sigmoid.forward(x)
correct_out = [
[0.18242552, 0.19836387, 0.21532798, 0.23332061, 0.25233322],
[0.27234476, 0.29332071, 0.31521239, 0.33795656, 0.36147556],
[0.3856778, 0.41045882, 0.4357028, 0.46128455, 0.48707185],
[0.51292815, 0.53871545, 0.5642972, 0.58954118, 0.6143222],
[0.63852444, 0.66204344, 0.68478761, 0.70667929, 0.72765524],
[0.74766678, 0.76667939, 0.78467202, 0.80163613, 0.81757448]
]
e = rel_error(correct_out, out)
print('Relative difference:', e)
self.assertTrue(e <= 5e-8)
def test_model(num_feat, num_classes):
classifier = ClassifierModel()
classifier.add_layer('FC-1', FCLayer(num_feat, 2))
classifier.add_layer('Sigmoid', Sigmoid())
classifier.add_layer('FC-2', FCLayer(2, 5))
classifier.add_layer('ReLU', ReLU())
classifier.add_layer('FC-3', FCLayer(5, 3))
classifier.add_layer('tanh', Tanh())
classifier.add_layer('FC-4', FCLayer(3, num_classes))
classifier.add_layer('Softmax', SoftmaxLayer())
return classifier
def test_model(num_feat, num_classes):
classifier = CNN_Classifier()
classifier.add_layer(
'Conv-1', ConvolutionLayer(num_feat, 2, kernel_size=3, stride=1,
pad=1))
classifier.add_layer('ReLU', ReLU())
classifier.add_layer(
'Conv-2', ConvolutionLayer(2, 3, kernel_size=3, stride=1, pad=1))
classifier.add_layer('tanh', Tanh())
classifier.add_layer(
'Conv-3', ConvolutionLayer(3, 3, kernel_size=3, stride=1, pad=0))
classifier.add_layer('Sigmoid', Sigmoid())
classifier.add_layer('Max-pool - 1',
MaxPoolingLayer(kernel_size=2, stride=1))
classifier.add_layer('FC-4', FCLayer(12, num_classes))
classifier.add_layer('Softmax', SoftmaxLayer())
return classifier
print()
print('===== Sigmoid Check =====')
"""
The results should be exactly same as below:
결과는 아래와 일치해야 합니다:
Forward:
[[0.26894142 0.88079708 0.62245933]
[0.5 0.47502081 0.52497919]]
Backward:
[[0. 0.10499359 0. ]
[0.25 0. 0.24937604]]
"""
sigmoid = Sigmoid()
temp4 = np.array([[-1, 2, 0.5], [0.0, -0.1, 0.1]])
temp5 = np.array([[0, 1, 0], [1, 0, 1]])
print('Forward: \n', sigmoid.forward(temp4))
print('Backward: \n', sigmoid.backward(temp5))
print()
print('===== InputLayer Check =====')
"""
The results should be exactly same as below:
결과는 아래와 일치해야 합니다:
Forward:
[[1.69405863 0. ]
[1.41846232 0. ]]
"""
# mnist / fashion_mnist
dataset = 'mnist'
# Hyper-parameters
num_epochs = 100
learning_rate = 0.001
reg_lambda = 1e-8
print_every = 10
batch_size = 128
# Add layers
model.add_layer('FC-1', FCLayer(784, 500))
model.add_layer('Sigmoid1', Sigmoid())
model.add_layer('FC-2', FCLayer(500, 500))
model.add_layer('Sigmoid2', Sigmoid())
model.add_layer('FC-3', FCLayer(500, 10))
model.add_layer('Softmax Layer', SoftmaxLayer())
# =========================================================================
assert dataset in ['mnist', 'fashion_mnist']
# Dataset
if dataset == 'fashion_mnist':
x_train, y_train, x_test, y_test = load_mnist('./data')
else:
x_train, y_train, x_test, y_test = load_fashion_mnist('./data')
