smlp

Classes

SMLP

Simple Multi-Layer Perceptron used in the project.

Module Contents

class smlp.SMLP(input_size=784, hidden_size=77, output_size=6)

Bases: torch.nn.Module

Simple Multi-Layer Perceptron used in the project.

The network has four linear layers. The first three are followed by ReLU activations and the final layer returns logits. The implementation expects flattened inputs of shape (batch_size, input_size) (for MNIST: input_size=784).

Parameters

input_sizeint: Dimensionality of the flattened input (default: 784 for 28x28 images).
hidden_sizeint: Number of neurons in the hidden layers (default: 77 in this project).
output_sizeint: Number of output classes (default: 6 for digits 4..9 remapped to 0..5).

layer1

layer2

layer3

layer4

relu

forward(x: torch.Tensor) → torch.Tensor

Forward pass.

Parameters

xtorch.Tensor: Input tensor of shape (batch_size, input_size). The function does not perform flattening; caller must supply a flattened tensor.

Returns

torch.Tensor: Logits tensor of shape (batch_size, output_size).