def test_tanh_1_forward(self):
print('\n==================================')
print(' Test tanh forward ')
print('==================================')
x = np.linspace(-1.5, 1.5, num=30).reshape(6, 5)
tanh = Tanh()
out = tanh.forward(x)
correct_out = [
[-0.90514825, -0.88460402, -0.85993717, -0.83047189, -0.79548953],
[-0.75425761, -0.70607365, -0.65032522, -0.58656543, -0.51459914],
[-0.43457022, -0.34703513, -0.25300497, -0.15393885, -0.05167806],
[0.05167806, 0.15393885, 0.25300497, 0.34703513, 0.43457022],
[0.51459914, 0.58656543, 0.65032522, 0.70607365, 0.75425761],
[0.79548953, 0.83047189, 0.85993717, 0.88460402, 0.90514825]
]
e = rel_error(correct_out, out)
print('Relative difference:', e)
self.assertTrue(e <= 5e-7)
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_tanh_2_backward(self):
print('\n==================================')
print(' Test tanh backward ')
print('==================================')
np.random.seed(123)
x = np.random.randn(5, 5)
tanh = Tanh()
out = tanh.forward(x)
dx = tanh.backward(x, 0.0)
correct_dx = [
[-0.39900528, 0.42055529, 0.26147548, -0.26911137, -0.42115391],
[0.22601191, -0.074565, -0.35876431, 0.34549378, -0.44238431],
[-0.44192875, -0.0938645, 0.27373408, -0.43556065, -0.36680073],
[-0.36170986, 0.10450339, 0.10753923, 0.41907822, 0.33386819],
[0.44697633, 0.27393871, -0.43260586, 0.37333295, -0.349298]
]
e = rel_error(correct_dx, dx)
print('Relative difference:', e)
self.assertTrue(e <= 5e-8)
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
"""
# 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('sigmoid-1', Tanh())
model.add_layer('FC-2', FCLayer(500, 500))
model.add_layer('sigmoid-2', Tanh())
model.add_layer('FC-3',FCLayer(500,10))
model.add_layer('Softmax Layer', SoftmaxLayer())
# =========================================================================
assert dataset in ['mnist', 'fashion_mnist']
# Dataset
if dataset == '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')
x_train, x_test = np.squeeze(x_train), np.squeeze(x_test)
# Hyper-parameters
num_epochs = 80
learning_rate = 0.1
reg_lambda = 1e-3
print_every = 10
batch_size = 128
# Add layers - example 1
#model.add_layer('FC-1', FCLayer(784, 10))
#model.add_layer('Sigmoid', Sigmoid())
#model.add_layer('Softmax Layer', SoftmaxLayer())
# Add layers - example 2
model.add_layer('FC-1', FCLayer(784, 200))
model.add_layer('Tanh1', Tanh())
model.add_layer('FC-2', FCLayer(200, 200))
model.add_layer('Tanh2', Tanh())
model.add_layer('FC-3', FCLayer(200, 10))
model.add_layer('Softmax Layer', SoftmaxLayer())
# =========================================================================
assert dataset in ['mnist', 'fashion_mnist']
# Dataset
if dataset == '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')
x_train, x_test = np.squeeze(x_train), np.squeeze(x_test)
