MCPcopy Create free account
hub / github.com/alexklwong/mondi-python / sparse_depth_error_weight

Function sparse_depth_error_weight

src/loss_utils.py:569–611  ·  view source on GitHub ↗

Given sparse depth, return tensor or integer evaluating teacher predictions against sparse depth Arg(s): sparse_depth : torch.Tensor[float32] N x 1 x H x W sparse depth tensor dilation_kernel_size : int Odd number of how large of a window to repeat s

(sparse_depth, dilation_kernel_size, teacher_outputs, w_sparse_error=1)

Source from the content-addressed store, hash-verified

source not stored for this graph (policy: none)

Callers

nothing calls this directly

Calls 1

dilate_sparse_depthFunction · 0.85

Tested by

no test coverage detected