Module conv2d 

Structs

Conv2dBackward

2-D convolution combined backward pass — computes both dx and dw in one kernel launch, using the same grid as the forward pass.

Conv2dBackwardDw

2-D convolution backward pass — gradient with respect to weights (dw).

Conv2dBackwardDx

2-D convolution backward pass — gradient with respect to input (dx).

Conv2dBiasForward

2-D convolution forward pass fused with a per-output-channel bias add.

Conv2dForward

2-D convolution forward pass (supports grouped and depthwise conv).

Conv2dOp

Functions

conv2d_backward

2-D convolution combined backward pass — computes both dx and dw in one kernel launch, using the same grid as the forward pass.

conv2d_backward_dw

2-D convolution backward pass — gradient with respect to weights (dw).

conv2d_backward_dx

2-D convolution backward pass — gradient with respect to input (dx).

conv2d_bias_forward

2-D convolution forward pass fused with a per-output-channel bias add.

conv2d_forward

2-D convolution forward pass (supports grouped and depthwise conv).