def test_manual_module(self):
np.random.seed(42)
torch.manual_seed(42)
for test_num in range(10):
n_batch = int(np.random.choice(range(32, 128)))
n_neurons = int(np.random.choice(range(1, 10)))
x = 2 * torch.randn(n_batch, n_neurons, requires_grad=True) + 10
bn_manual_mod = CustomBatchNormManualModule(n_neurons)
y_manual_mod = bn_manual_mod(x)
self.assertLess(np.max(np.abs(y_manual_mod.mean(dim=0).data.numpy())), 1e-5)
self.assertLess(np.max(np.abs(y_manual_mod.var(dim=0).data.numpy() - 1)), 1e-1)
def __init__(self, n_inputs, n_hidden, n_classes):
"""
Initializes MLP object.
Args:
n_inputs: number of inputs.
n_hidden: list of ints, specifies the number of units
in each linear layer. If the list is empty, the MLP
will not have any linear layers, and the model
will simply perform a multinomial logistic regression.
n_classes: number of classes of the classification problem.
This number is required in order to specify the
output dimensions of the MLP
Implement initialization of the network.
"""
########################
# PUT YOUR CODE HERE #
#######################
super(MLP, self).__init__()
self.layers = []
in_features = n_inputs
for out_features in n_hidden:
linear = nn.Linear(in_features, out_features)
##batchnorm = nn.BatchNorm1d(out_features)
# dropout = nn.Dropout(0.2)
relu = nn.ReLU()
self.layers.append(linear)
##self.layers.append(batchnorm)
self.layers.append(CustomBatchNormManualModule(out_features))
# self.layers.append(dropout)
self.layers.append(relu)
in_features = out_features
#dropout = nn.Dropout()
#self.layers.append(dropout)
linear = nn.Linear(in_features, n_classes)
softmax = nn.Softmax()
self.layers.append(linear)
# self.layers.append(softmax)
self.sequential = nn.Sequential(*self.layers)