What is nn.Module and how do you build a custom neural network in PyTorch?

nn.Module is the base class for all neural network components in PyTorch. Subclassing it gives you parameter management, device placement, train/eval mode toggling, state dict serialisation, and hooks — all for free.

import torch
import torch.nn as nn

class MLP(nn.Module):
    def __init__(self, in_features: int, hidden: int, out_features: int):
        super().__init__()   # MUST call this first!

        # Layers defined as attributes are auto-registered as sub-modules
        self.fc1  = nn.Linear(in_features, hidden)
        self.relu = nn.ReLU()
        self.drop = nn.Dropout(p=0.3)
        self.fc2  = nn.Linear(hidden, out_features)

    def forward(self, x: torch.Tensor) -> torch.Tensor:
        """Define the forward computation."""
        x = self.fc1(x)
        x = self.relu(x)
        x = self.drop(x)
        x = self.fc2(x)
        return x

# Instantiate and inspect
model = MLP(in_features=784, hidden=256, out_features=10)

# Forward pass — calls forward() via __call__
x = torch.randn(32, 784)   # batch of 32
out = model(x)              # shape (32, 10)

# Parameter inspection
for name, param in model.named_parameters():
    print(name, param.shape, param.requires_grad)
# fc1.weight  torch.Size([256, 784])  True
# fc1.bias    torch.Size([256])       True
# fc2.weight  torch.Size([10, 256])   True
# fc2.bias    torch.Size([10])        True

total = sum(p.numel() for p in model.parameters())
print(f"Total parameters: {total:,}")

Critical rules:

• Always call super().__init__() in __init__
• Define layers as attributes (not local variables) so PyTorch registers them
• Implement the forward() method — never call it directly; use model(x) which invokes hooks
• Use model(x) not model.forward(x) so pre/post-forward hooks fire

Invest now in Acorns!!! 🚀
Join Acorns and get your $5 bonus!

Acorns is a micro-investing app that automatically invests your "spare change" from daily purchases into diversified, expert-built portfolios of ETFs. It is designed for beginners, allowing you to start investing with as little as $5. The service automates saving and investing. Disclosure: I may receive a referral bonus.

Invest now!!! Get Free equity stock (US, UK only)!

Use Robinhood app to invest in stocks. It is safe and secure. Use the Referral link to claim your free stock when you sign up!.

The Robinhood app makes it easy to trade stocks, crypto and more.

Webull! Receive free stock by signing up using the link: Webull signup.