oa::ml::nn::Conv2d
Trainable grouped FP32 two-dimensional convolution over NCHW matrices.
Declaration
pub struct Conv2d { /* private fields */ }Public Rust declaration.
Public Methods
Associated Functions
pub fn new_with_seed( input_channels: usize, output_channels: usize, kernel_size: usize, stride: usize, padding: usize, groups: usize, seed: u64, ) -> Result<Self>Public Method Documentation
No public source comment is attached to this declaration.
No public source comment is attached to this declaration.
No public source comment is attached to this declaration.
No public source comment is attached to this declaration.
No public source comment is attached to this declaration.
No public source comment is attached to this declaration.
No public source comment is attached to this declaration.
No public source comment is attached to this declaration.
No public source comment is attached to this declaration.
No public source comment is attached to this declaration.
No public source comment is attached to this declaration.
No public source comment is attached to this declaration.
No public source comment is attached to this declaration.
No public source comment is attached to this declaration.
No public source comment is attached to this declaration.
No public source comment is attached to this declaration.
No public source comment is attached to this declaration.
No public source comment is attached to this declaration.
No public source comment is attached to this declaration.
No public source comment is attached to this declaration.
No public source comment is attached to this declaration.
No public source comment is attached to this declaration.
No public source comment is attached to this declaration.
No public source comment is attached to this declaration.
No public source comment is attached to this declaration.
No public source comment is attached to this declaration.
Construct a deterministically initialized grouped Conv2d layer. The weight initializer is symmetric uniform with bound 1 / sqrt((input_channels / groups) * kernel_size²); bias starts at zero. Errors Returns an error when dimensions are zero, channels are not divisible by groups, shape arithmetic overflows, or allocation/upload fails.
No public source comment is attached to this declaration.