Skip to main content

🏗️ Building Models

In PyTorch, every model you build must inherit from nn.Module.

🧱 The Blueprint​

You always need two things:

  1. __init__(): Where you declare all your layers (your ingredients).
  2. forward(): Where you explain how the data flows through those layers (the recipe).

🐍 Python Implementation​

Let's combine everything we've learned (Linear Layers, ReLU, Dropout) into a clean class!

import torch
import torch.nn as nn

class MyAwesomeModel(nn.Module):
    def __init__(self):
        # Always call super() first!
        super().__init__()
        
        # Declare Ingredients
        self.fc1 = nn.Linear(20, 64)
        self.relu = nn.ReLU()
        self.dropout = nn.Dropout(0.2)
        self.fc2 = nn.Linear(64, 2)
        
    def forward(self, x):
        # Follow the Recipe
        x = self.fc1(x)
        x = self.relu(x)
        x = self.dropout(x)
        x = self.fc2(x)
        return x

model = MyAwesomeModel()
print(model) # PyTorch beautifully prints your architecture!