Struct Conv1d 

pub struct Conv1d { /* private fields */ }

1D Convolution layer.

Applies a 1D convolution over an input signal composed of several input planes.

Shape

• Input: (N, C_in, L) where N is batch size, C_in is input channels, L is length
• Output: (N, C_out, L_out) where L_out = (L + 2*padding - kernel_size) / stride + 1

Example

use aprender::nn::{Conv1d, Module};
use aprender::autograd::Tensor;

let conv = Conv1d::new(16, 32, 3);  // 16 in channels, 32 out channels, kernel size 3
let x = Tensor::randn(&[4, 16, 100]);  // batch of 4, 16 channels, length 100
let y = conv.forward(&x);  // [4, 32, 98]

Implementations

impl Conv1d

pub fn new(in_channels: usize, out_channels: usize, kernel_size: usize) -> Self

Create a new Conv1d layer.

Arguments

• in_channels - Number of input channels
• out_channels - Number of output channels
• kernel_size - Size of the convolving kernel

pub fn with_options( in_channels: usize, out_channels: usize, kernel_size: usize, stride: usize, padding: usize, bias: bool, ) -> Self

Create Conv1d with custom options.

Arguments

• in_channels - Number of input channels
• out_channels - Number of output channels
• kernel_size - Size of the convolving kernel
• stride - Stride of the convolution
• padding - Zero-padding added to both sides
• bias - If true, adds a learnable bias

pub fn with_layout( in_channels: usize, out_channels: usize, kernel_size: usize, stride: usize, padding: usize, bias: bool, layout: ConvLayout, ) -> Self

Create Conv1d with a specific data layout.

Arguments

• in_channels - Number of input channels
• out_channels - Number of output channels
• kernel_size - Size of the convolving kernel
• stride - Stride of the convolution
• padding - Zero-padding added to both sides
• bias - If true, adds a learnable bias
• layout - Data layout for input/output tensors

pub fn with_stride( in_channels: usize, out_channels: usize, kernel_size: usize, stride: usize, ) -> Self

Create Conv1d with specific stride.

pub fn with_padding( in_channels: usize, out_channels: usize, kernel_size: usize, padding: usize, ) -> Self

Create Conv1d with padding.

pub fn kernel_size(&self) -> usize

Get kernel size.

pub fn stride(&self) -> usize

Get stride.

pub fn padding(&self) -> usize

Get padding.

Trait Implementations

impl Debug for Conv1d

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Module for Conv1d

fn forward(&self, input: &Tensor) -> Tensor

Perform forward computation.

fn parameters(&self) -> Vec<&Tensor>

Get references to all learnable parameters.

fn parameters_mut(&mut self) -> Vec<&mut Tensor>

Get mutable references to all learnable parameters.

fn refresh_caches(&mut self)

Refresh any cached computations after parameters have been modified.

fn train(&mut self)

Set the module to training mode.

fn eval(&mut self)

Set the module to evaluation mode.

fn training(&self) -> bool

Check if the module is in training mode.

fn zero_grad(&mut self)

Zero out gradients for all parameters.

fn num_parameters(&self) -> usize

Get the number of learnable parameters.