Skip router auxiliary loss compute when their lambda is 0 (gitea #31)
CI / Format (ruff format) (push) Successful in 42s
CI / Lint (ruff check) (push) Successful in 44s
CI / Sync project version with tag (push) Has been skipped
CI / Lint (ruff check) (pull_request) Successful in 33s
CI / Type check (ty) (push) Successful in 37s
CI / Format (ruff format) (pull_request) Successful in 32s
CI / Sync project version with tag (pull_request) Has been skipped
CI / Type check (ty) (pull_request) Successful in 32s
CI / Tests (pull_request) Successful in 3m55s
CI / Tests (push) Successful in 3m57s

FlowDDPMStageTrainer._compute unconditionally called
router.balance_loss/classify_loss/entropy_loss whenever a router existed,
then only added each term into total if its lambda was > 0 -- so every
routed run paid for balance_loss/entropy_loss's extra router.gate(...)
forward passes even at the default lambda_balance = lambda_proc =
lambda_entropy = 0.0 (the exact config the failed 2026-07-22 router
benchmark ran). Guard each computation on the same > 0 condition that
already guarded the addition, matching WGANStageTrainer's cost structure
which has no router-loss block at all. total's value is unchanged either
way. Added a test that spies on the router's three loss methods and
checks call counts both at lambda=0 (must be skipped) and lambda>0 (must
still run, so the guard doesn't suppress the real path).

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
This commit is contained in:
2026-08-13 16:18:19 +02:00
parent 32aa5a5f92
commit f505fe7f22
2 changed files with 61 additions and 3 deletions
+6 -3
View File
@@ -618,9 +618,12 @@ class FlowDDPMStageTrainer(StageTrainer):
l_balance = l_proc = l_entropy = torch.zeros((), device=device) l_balance = l_proc = l_entropy = torch.zeros((), device=device)
if self.router is not None: if self.router is not None:
l_balance = self.router.balance_loss(cond_cont, cond_cat) if self.spec.lambda_balance > 0:
l_proc = self.router.classify_loss(cond_cont, cond_cat, proc_idx) l_balance = self.router.balance_loss(cond_cont, cond_cat)
l_entropy = self.router.entropy_loss(cond_cont, cond_cat) if self.spec.lambda_proc > 0:
l_proc = self.router.classify_loss(cond_cont, cond_cat, proc_idx)
if self.spec.lambda_entropy > 0:
l_entropy = self.router.entropy_loss(cond_cont, cond_cat)
total = self.spec.lambda_weight * l_gen + self.spec.n_sec_lambda * l_nsec + self.particle_type_lambda * l_type total = self.spec.lambda_weight * l_gen + self.spec.n_sec_lambda * l_nsec + self.particle_type_lambda * l_type
if self.spec.lambda_balance > 0: if self.spec.lambda_balance > 0:
+55
View File
@@ -5,6 +5,7 @@ import csv
import math import math
import tempfile import tempfile
from pathlib import Path from pathlib import Path
from unittest.mock import MagicMock
import pytest import pytest
import torch import torch
@@ -587,6 +588,60 @@ def test_stage_spec_from_config_omitted_decoder_and_particle_type_match_default_
assert spec.particle_type.target == "onehot" assert spec.particle_type.target == "onehot"
def _routed_stage1_trainer(lambda_balance, lambda_proc, lambda_entropy):
cfg = _base_cfg()
cfg["stage1_model"]["router"] = {
"enabled": True,
"type": "energy",
"n_experts": 3,
"temperature": 0.5,
"learn_centers": True,
"lambda_balance": lambda_balance,
"lambda_proc": lambda_proc,
"lambda_entropy": lambda_entropy,
}
model_config = _model_config(cfg)
models = build_models(model_config)
critics = build_critics(model_config)
trainers = build_stage_trainers(cfg, models, critics, torch.device("cpu"), total_train_batches=4)
return trainers["stage1"]
def test_router_aux_losses_skipped_when_lambda_zero_but_run_when_positive():
"""Gitea #31: FlowDDPMStageTrainer._compute must not call
router.balance_loss/classify_loss/entropy_loss when the corresponding
lambda is 0 (the default) -- those calls do their own router.gate(...)
forward pass that is wasted once the term is masked out of the total
loss anyway. Checked both ways: zero lambdas must skip all three calls,
positive lambdas must still make them (the guard must not accidentally
suppress the real path)."""
batch = _fake_batches(1, 4)[0]
device = torch.device("cpu")
trainer_zero = _routed_stage1_trainer(0.0, 0.0, 0.0)
router_zero = trainer_zero.router
router_zero.balance_loss = MagicMock(wraps=router_zero.balance_loss)
router_zero.classify_loss = MagicMock(wraps=router_zero.classify_loss)
router_zero.entropy_loss = MagicMock(wraps=router_zero.entropy_loss)
stats_zero = trainer_zero.step(batch, device, global_step=1)
assert router_zero.balance_loss.call_count == 0
assert router_zero.classify_loss.call_count == 0
assert router_zero.entropy_loss.call_count == 0
assert stats_zero["loss_balance"] == 0.0
assert stats_zero["loss_proc"] == 0.0
assert stats_zero["loss_entropy"] == 0.0
trainer_pos = _routed_stage1_trainer(0.1, 0.1, 0.01)
router_pos = trainer_pos.router
router_pos.balance_loss = MagicMock(wraps=router_pos.balance_loss)
router_pos.classify_loss = MagicMock(wraps=router_pos.classify_loss)
router_pos.entropy_loss = MagicMock(wraps=router_pos.entropy_loss)
trainer_pos.step(batch, device, global_step=1)
assert router_pos.balance_loss.call_count == 1
assert router_pos.classify_loss.call_count == 1
assert router_pos.entropy_loss.call_count == 1
# --- AR trainer wiring (v0.3.0 step 5) -------------------------------------- # --- AR trainer wiring (v0.3.0 step 5) --------------------------------------