type

benchmarks.hota.learned_modules.Propagator

class Propagator(nn.Module):

Predict step: residual MLP over a track embedding, renormalized.

Members

method

  • __init__Build a ``dim -> hidden -> dim`` residual MLP.
  • forwardPropagate ``x`` ``(N, D)`` one step and renormalize to unit norm.

property

Source: unitrack/benchmarks/hota/learned_modules.py:26