def __init__(self,
size: int,
activation_function: ActivationFunction = ReLu(),
optimizer: Optimizer = Adam(),
name="Simple Layer"):
self.name = name
self.size = size
self._activation_function = activation_function
self._initialised = False
# Unknown for now, gets updated then when self.initialize get called.
self._previous_layer_size = 0
# Parameters for the affine transformation from layer l-1 to layer l
self.W = None
self.b = np.zeros(
size) + 0.001 # Initialization with a small magnitude
# Last activation from previous layer
self.a_lm1 = None
# Last affine transformation of combination of inputs on this layer
self.z_l = None
# Error on this layer l
self.delta_l = None
# Last gradient
self._last_d_W = 0
self._last_d_b = 0
# Optimizers for weights and biases
self._W_optimizer = copy.deepcopy(optimizer)
self._b_optimizer = copy.deepcopy(optimizer)
def __init__(self, size: int, activation_function: ActivationFunction = ReLu(), name="Simple Layer"):
self.name = name
self.size = size
self._activation_function = activation_function
self._initialised = False
# Unknown for now, gets updated then when self.initialize get called.
self._previous_layer_size = 0
# Parameters for the affine transformation from layer l-1 to layer l
self.W = None
self.b = np.zeros(size) + 0.001 # Initialization with a small magnitude
# Last activation from previous layer
self.a_lm1 = None
# Last affine transformation of combination of inputs on this layer
self.z_l = None
# Error on this layer l
self.delta_l = None
# Last gradient
self._last_d_W = 0
self._last_d_b = 0
# Adam parameters
self.eps = 10 ** (-8)
self.t = 1
self.beta1_w = 0.9
self.beta2_w = 0.999
self.m_w = 0
self.v_w = 0
self.beta1_b = 0.9
self.beta2_b = 0.999
self.m_b = 0
self.v_b = 0
def __init__(self, size: int):
super().__init__(size, activation_function=ReLu(), name="SimpleMSEOutputLayer")
y_test = y_test.astype(int)
return x_train, y_train, x_test, y_test
if __name__ == "__main__":
# Loading data
data_folder = "../data" # set your own value
x_train, y_train, x_test, y_test = load_sets(data_folder)
y_train = one_hot(y_train)
y_test = one_hot(y_test)
# Defining the network
network = Network(input_size=14, name="14-100-40-4 Arch")
network.stack(Layer(size=100, activation_function=ReLu()))
network.stack(Layer(size=40, activation_function=ReLu()))
network.output(SoftMaxCrossEntropyOutputLayer(size=4))
# Printing information
print(network)
# Defining a log file
current_datetime = strftime("%Y-%m-%d-%H:%M:%S", gmtime())
logs_folder = os.path.join("..", "logs")
out_file = os.path.join(
logs_folder, f"benchmark_{network.name}-{current_datetime}.csv")
# Benchmarking the network
logger = network.benchmark(x_train,