ApplyKind#

class t3toolbox.backend.fitting.ApplyKind#

Bases: _WwSample, ScalarOutputKind

The all-modes apply: one scalar per measurement. Unweighted (no mode or order axis).

name = 'apply'#

Methods#

precompute(frame, ww)

The reusable frame sweep at frame for sample (computed once per local model).

forward(v, ww, frame, sweep)

The bare Jacobian J v (no gauge).

transpose(r, ww, frame, sweep)

The bare transpose J^T r, summed over the sample stack W; raw (dU, dG).

point_forward(x_cores, ww)

The point operation S(x) -- what the residual is measured against.

take(sample, data, idx)

Gather a flat W subset of (sample, data) -- the default minibatch draw's layout hook.