Classes and Objects

A class is a blueprint; an object is an instance created from that blueprint.

class Model:
    def __init__(self, name):
        self.name = name
        self.accuracy = 0.0

    def train(self):
        print(f"Training {self.name}...")
        self.accuracy = 0.95

    def evaluate(self):
        print(f"Accuracy: {self.accuracy}")

my_model = Model("NeuralNet")
my_model.train()
my_model.evaluate()

• Encapsulation: Keep data (weights) and methods (forward, backward) together.
• Inheritance: Create specialized model classes from a base class.
• Frameworks: TensorFlow and PyTorch use classes for layers, optimizers, and datasets.

class DenseLayer:
    def __init__(self, input_size, output_size):
        self.weights = [[0]*output_size for _ in range(input_size)]

    def forward(self, inputs):
        # simplified matrix multiplication
        return inputs

• Class = blueprint; object = actual instance.
• __init__ is the constructor (sets up initial state).
• Methods are functions inside a class.
• OOP helps manage complexity in AI projects.