import nnfs

from nnfs.datasets import spiral_data

class Layer_Dense:

def __init__(self,n_inputs,n_neurons):

self.weights = 0.01*np.random.randn(n_inputs, n_neurons) #generating random data for weights

self.biases = np.zeros((1,n_neurons)) #Biases data

def forward(self, inputs):

self.output = np.dot(inputs, self.weights) + self.biases #inputs from last layer to the next layer

class Activation_ReLU:

def forward(self,inputs):

self.output = np.maximum(0,inputs)

from nnfs.datasets import spiral_data

X,y = spiral_data(samples = 100, classes = 3)

dense1 = Layer_Dense(2,3)

activation1 = Activation_ReLU()

dense1.forward(X)

activation1.forward(dense1.output)

print(activation1.output[:5])