Gradient of batched block_diag falls back to numba object mode (Blockwise ARange)

Aperta
#2,702 1 commento 0 reazioni 0 assegnatari Vedi su GitHub

I maintainer di solito rispondono entro 1 giorno

@saksham10arora-dotcom ci sta già lavorando.

Dal 21/11/2026.

  • #2703 di @saksham10arora-dotcom — aperta

Valutazione

Difficoltà
4/5
Tempo stimato
3-5 giorni
Idoneità per principianti
35/100
Tipo di issue
Bug
Chiarezza
Abbastanza chiara
Stato di attività
Attiva
Stack tecnologico
numpy, python
Ambito
performance

Direzione di ricerca

Reproduce with the snippet in mode="NUMBA" and confirm the Blockwise{ARange} object-mode warning and the timing gap. Then read BlockDiagonal.pullback and how Blockwise calls vectorize_shape, since that is where core shapes become batched; compare against the concatenate-based construction to see what a non-batched pullback would look like. Done means the gradient runs in nopython mode, timings approach the concatenate figure, and a regression test covers batched plus unbatched inputs.

Scritto dal modello di indicizzazione a partire dal testo della issue.

Descrizione

When one input to block_diag is batched, its gradient compiles a Blockwise{ARange} that numba runs in object mode. In a small example the gradient is about 20x slower than building the same matrix with concatenate.

import pytensor
import pytensor.tensor as pt

A = pt.tensor3("A")  # batch of 6x6 blocks
B = pt.matrix("B")   # 2x2, not batched
out = (pt.linalg.block_diag(A, B) ** 2).sum()
f = pytensor.function([A, B], pytensor.grad(out, [A, B]), mode="NUMBA")
# UserWarning: Numba will use object mode to run Blockwise{ARange{dtype='int64'}, (),(),()->(o00)}'s perform method

Timings with A of shape (1000, 6, 6), pytensor 3.3.8, M-series Mac:

gradient of ms per call
block_diag(A, B) as it is now 3.2
same, with the pullback using slices instead of arange 1.1 (still object mode, now Blockwise{Subtensor})
the same matrix built with concatenate and zeros 0.14

What I think happens: BlockDiagonal.pullback indexes gout with arange(index_begin, index_end), computed from i.shape. When Blockwise vectorizes the pullback, vectorize_shape broadcasts each core shape over the batch (which is correct in general), so the indices and the arange become batched even though they are identical for every batch element. Swapping arange for slices removes the ARange, but the slice bounds are still batched, so it moves the problem to Blockwise{Subtensor}.

Where I hit it: pymc-extras structural models. TimeSeasonality(duration > 1) has a time-varying transition matrix, and adding any other component (e.g. Autoregressive) combines them with block_diag(T, o_T), so the Kalman filter gradient picks this up (pymc-devs/pymc-extras#833).

I'm not sure what the preferred fix is: a pullback that avoids batched indices, a rewrite that drops batch dims from Blockwise inputs that only come from broadcast shapes, or something else. Happy to work on it with some direction.

Lingua principale
Python
Stelle
654
Fork
258
Merge medio
2g 7h
PR unite (30g)
39

Preparare l'ambiente

Come iniziare

  1. Leggi tutta la issue e poi la guida ai contributi del progetto.
  2. Commenta sulla issue per dire che te ne occupi tu — evita che due persone facciano lo stesso lavoro.
  3. Fai un fork del repository e lavora su un branch.
  4. Apri una pull request che faccia riferimento al numero della issue.

Altre issue di pymc-devs/pytensor

Tutte le issue di pymc-devs/pytensor

Issue simili

Altre issue su Python

Ricevi le nuove issue nella tua casella

Un breve riepilogo di issue GitHub adatte ai principianti.