Build Your First Neural Network

Learn to build, train, and evaluate a simple feedforward neural network from scratch using PyTorch in DLWΛY.

In this tutorial, you'll build a neural network that classifies handwritten digits from the MNIST dataset.

Prerequisites

Basic Python knowledge
A DLWΛY account
A Jupyter or Kaggle kernel attached under Settings → Compute. PyTorch runs there; the in-browser Python runtime covers NumPy, pandas and scikit-learn

Step 1: Open DLWΛY

Open the Studio, create a project, and open the Explorer. Create a Python file for the model.

Step 2: Define the Model

python

import torchimport torch.nn as nn class NeuralNet(nn.Module):    def __init__(self):        super(NeuralNet, self).__init__()        self.flatten = nn.Flatten()        self.layers = nn.Sequential(            nn.Linear(784, 512),            nn.ReLU(),            nn.Dropout(0.2),            nn.Linear(512, 256),            nn.ReLU(),            nn.Dropout(0.2),            nn.Linear(256, 10)        )     def forward(self, x):        x = self.flatten(x)        return self.layers(x) model = NeuralNet()print(model)

Step 3: Train the Model

python

criterion = nn.CrossEntropyLoss()optimizer = torch.optim.Adam(model.parameters(), lr=0.001) for epoch in range(10):    for images, labels in train_loader:        outputs = model(images)        loss = criterion(outputs, labels)        optimizer.zero_grad()        loss.backward()        optimizer.step()    print(f'Epoch {epoch+1}/10, Loss: {loss.item():.4f}')

Step 4: Evaluate

Run the model on the test set and see your accuracy, typically 98%+ after 10 epochs.

Next Steps

Try changing the hidden layer sizes
Add batch normalization
Experiment with different optimizers