Sequential

Sequential chains a list of Module layers and calls them in order during the forward pass, passing the output of each layer as the input to the next. It is the fastest way to build simple feed-forward architectures without writing a custom Module subclass. All nested parameters are discovered automatically and can be passed to an optimizer.

import simplegrad as sg
import simplegrad.nn as nn

model = nn.Sequential(
    nn.Linear(784, 128),
    nn.ReLU(),
    nn.Linear(128, 10),
)
out = model(sg.normal((32, 784)))

Sequential

Bases: Module

A sequential container that applies modules in the order they are passed.

Parameters:

• *modules (Module, default: () ) –

  Any number of Module instances to chain together.

Attributes

Attribute	Type	Description
.modules	list[Module]	Ordered list of layers passed at construction.

Methods

Method	Description
.forward()	Pass x through each layer in order and return the final output.

Inherits all methods from Module: .parameters(), .submodules(), .to_device(), .summary(), .set_train_mode(), .set_eval_mode().