from nnfs.datasets import spiral_data from libraries import Activation_Softmax_Loss_CategoricalCorssentropy from libraries import Activation_Softmax from libraries import Loss_CategoricalCrossentropy from libraries import Activation_ReLU from libraries import Layer_Dense # initialize nnfs dataset nnfs.init() # Create dataset x, y = spiral_data(samples=100, classes=3) # Create Dense Layer with 2 input features and 3 output values dense1 = Layer_Dense(2, 3) # Create ReLU activation (to be used with Dense Layer) activation1_relu = Activation_ReLU() # Create second Dense layer with 3 input features (as we take ouput of previous layer here) # and 3 output values (output values) dense2 = Layer_Dense(3, 3) # Create a softmax classfier's combined loss and activation loss_activation = Activation_Softmax_Loss_CategoricalCorssentropy() # Perform a forward pass of our training data through this layer dense1.forward(x) # Perform a forward pass through activation function
import numpy as np import nnfs from nnfs.datasets import spiral_data from libraries import Layer_Dense from libraries import Activation_ReLU from libraries import Activation_Softmax # Initializes NNFS nnfs.init() # Create dataset x,y = spiral_data(samples=100, classes=3) # Create Dense layer with 2 input features and 3 output values dense1 = Layer_Dense(2,3) # Create Relu Activation (to be used with Dense Layer) activation_relu = Activation_ReLU() # Create second Dense layer with 3 input features (as we take output of previous layer here) # and 3 output values dense2 = Layer_Dense(3, 3) # Create Softmax actiavtion (to be used with Dense layer): activation_softmax = Activation_Softmax() # Make a forward pass of our training data through this layer dense1.forward(x) # Make a forward pass through activation fucntion
def __init__(self, learning_rate=1.0):
self.learning_rate = learning_rate
# Update parameters
def update_params(self, layer):
# Our goal to minimize value and gradient point towards steepest function ascent,
# thus we substractd learning rate
layer.weights += -self.learning_rate * layer.dweights
layer.biases += -self.learning_rate * layer.dbiases
# create dataset, with 100 sample 3 features
x, y = spiral_data(samples=100, classes=3)
# Create Dense layer with 2 input fatures and 64 output values
dense1 = Layer_Dense(2, 64)
# Create Relu activation (to be used with Dense Layer)
activation_relu = Activation_ReLU()
# Create Dense layer with 2 input fatures and 64 output values
dense_hidden = Layer_Dense(64, 64)
# Create Relu activation (to be used with Dense Layer)
activation_hidden_relu = Activation_ReLU()
# Create second Dense layer with 64 input features (as we take output of
# previous layer here) and 3 output values (output values)
dense2 = Layer_Dense(64, 3)
# create softmax classfier's combined loss and activation
from libraries import Layer_Dense from libraries import Loss_CategoricalCrossentropy from libraries import Activation_ReLU from libraries import Activation_Softmax # Optimization determine how to adjust the weights and biases to decrease the loss. # Finding an intelligent way to adjust the neurons’ input’s weights and biases to minimize loss is the main difficulty of neural networks. # initialize dataset nnfs.init() # Create dataset x, y = vertical_data(samples=100, classes=3) # Create model dense1 = Layer_Dense(2, 3) # First dense layer with 2 input, 3 output activation_relu = Activation_ReLU() dense2 = Layer_Dense(3, 3) # Second dense layer, 3 inputs, 3 output activation_softmax = Activation_Softmax() # Create loss function loss_function = Loss_CategoricalCrossentropy() # Then create some variables to track the best loss and the associated weights and biases lowest_loss = 9999999 # some initial value best_dense1_weights = dense1.weights.copy() best_dense1_biases = dense1.biases.copy() best_dense2_weights = dense2.weights.copy() best_dense2_biases = dense2.biases.copy() # we iterate as many times as desired, pick random values for weights and biases, and save the weights
# in plain python
# output = []
# for i in input:
# val = 0 if x <= 0 else x
# output.append(val)
# self.output = output
# use numpy to create more clear result
# Calculate ouput values from input
self.output = np.maximum(0, input)
# Create datasets
x, y = spiral_data(100, 3)
# Create Dense layer with 2 input features and 3 output layer
dense1 = Layer_Dense(2, 3)
# Create ReLU activation (to be used with Dense Layer)
activation1 = Activation_ReLU()
# Make a forward pass of our training data through this layer
dense1.forward(x)
# Forward pass through activation func.
# Takes in output from previous layer
activation1.forward(dense1.output)
# Print first 5 rows
print(activation1.output[:5])
def post_update_params(self):
self.iterations += 1
# --------------------------------
# initialize nnfs dataset
nnfs.init()
# Create dataset
x, y = spiral_data(samples=100, classes=3)
# Create dense layer with 2 input features and 64 output valus
dense1 = Layer_Dense(2, 64)
# Create softmax classfier's combined loss and activation
loss_activation = Activation_Softmax_Loss_CategoricalCorssentropy()
optimizer = Optimizer_Adam(learning_rate=0.02, decay=1e-5)
# forward
dense1.forward(x)
loss_activation.forward(dense1.output,y)
dense1.backward(loss_activation.output)
# Update weight and biases
optimizer.pre_update_params()