5. Creating a Simple Neural Network

Introduction

I've heard about neural networks in programming. How can I create a simple one? 🧠

Neural networks are the foundation of deep learning. Let's create a simple one using Python and PyTorch!

Sounds fun! Let's get started! 🚀

Step 1: Import Libraries

What libraries do we need?

We'll use PyTorch and torchvision for this example.

import torch
import torch.nn as nn
import torchvision

Step 2: Define the Neural Network

Now, let's create the neural network!

We'll create a simple feedforward neural network with one hidden layer.

class SimpleNeuralNetwork(nn.Module):
    def __init__(self, input_size, hidden_size, output_size):
        super(SimpleNeuralNetwork, self).__init__()
        self.fc1 = nn.Linear(input_size, hidden_size)
        self.relu = nn.ReLU()
        self.fc2 = nn.Linear(hidden_size, output_size)

    def forward(self, x):
        out = self.fc1(x)
        out = self.relu(out)
        out = self.fc2(out)
        return out

Step 3: Create an Instance of the Neural Network

How do we use our neural network?

First, we create an instance with the desired input, hidden, and output sizes.

input_size = 784
hidden_size = 100
output_size = 10

model = SimpleNeuralNetwork(input_size, hidden_size, output_size)

Step 4: Define Loss and Optimizer

What's next?

We need to define a loss function and an optimizer to train the network.

loss_function = nn.CrossEntropyLoss()
optimizer = torch.optim.Adam(model.parameters(), lr=0.001)

Step 5: Train the Neural Network

Time for training! 🏋️‍♀️

We'll use a dataset, like MNIST, to train our neural network. For simplicity, we won't include the dataset loading code here.

num_epochs = 5

for epoch in range(num_epochs):
    for i, (images, labels) in enumerate(train_loader):
        images = images.view(-1, 28*28)

        # Forward pass
        outputs = model(images)
        loss = loss_function(outputs, labels)

        # Backward pass
        optimizer.zero_grad()
        loss.backward()
        optimizer.step()

Conclusion

Great job! You’ve created a simple neural network using Python and PyTorch. We defined the network, created an instance, defined loss and optimizer, and trained the network. Keep exploring and have fun with neural networks! 🎉