Mojo module

normalization

Functions

block_reduce:
group_norm:
group_norm_gpu:
group_norm_gpu_block:
group_norm_gpu_warp_tiling:
group_norm_reshape: Reshapes an input buffer for group normalization by flattening all dimensions except the group dimension. Returns a 2D buffer of shape (num_groups * N, group_size), where group_size is the product of channels_per_group and spatial.
group_norm_shape:
layer_norm:
layer_norm_cpu: Computes layernorm(elementwise_fn(x)) across the last dimension of x, where layernorm is defined as $(x-mean(x))/(sqrt(var(x)+eps)*gamma_fn + beta$ .
layer_norm_gpu:
layer_norm_gpu_block:
layer_norm_gpu_warp_tiling:
layer_norm_reshape:
layer_norm_shape: Compute the output shape of a layer_norm operation.
rms_norm:
rms_norm_cpu:
rms_norm_gpu:
rms_norm_gpu_block:
rms_norm_gpu_warp_tiling:
rms_norm_shape:
welford_block_all_reduce:
welford_combine:
welford_update:
welford_warp_all_reduce:
welford_warp_reduce: