ModuFlow/mask-ddpm
3
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

update

Browse Source
This commit is contained in:
MingzheYang
2026-01-27 00:04:54 +08:00
parent 3a1ad6b206
commit e09695319c
2 changed files with 29 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
example/config.json
View File

@@ -46,6 +46,9 @@
"temporal_lr": 0.001, "temporal_lr": 0.001,
"quantile_loss_weight": 0.2, "quantile_loss_weight": 0.2,
"quantile_points": [0.05, 0.25, 0.5, 0.75, 0.95], "quantile_points": [0.05, 0.25, 0.5, 0.75, 0.95],
"snr_weighted_loss": true,
"snr_gamma": 1.0,
"residual_stat_weight": 0.05,
"sample_batch_size": 8, "sample_batch_size": 8,
"sample_seq_len": 128 "sample_seq_len": 128
} }

26
example/train.py
View File

@@ -73,6 +73,9 @@ DEFAULTS = {
"temporal_lr": 1e-3, "temporal_lr": 1e-3,
"quantile_loss_weight": 0.0, "quantile_loss_weight": 0.0,
"quantile_points": [0.05, 0.25, 0.5, 0.75, 0.95], "quantile_points": [0.05, 0.25, 0.5, 0.75, 0.95],
"snr_weighted_loss": True,
"snr_gamma": 1.0,
"residual_stat_weight": 0.0,
} }
@@ -329,6 +332,13 @@ def main():
loss_cont = (loss_base * weights).mean() loss_cont = (loss_base * weights).mean()
else: else:
loss_cont = loss_base.mean() loss_cont = loss_base.mean()
if bool(config.get("snr_weighted_loss", False)):
a_bar_t = alphas_cumprod[t].view(-1, 1, 1)
snr = a_bar_t / torch.clamp(1.0 - a_bar_t, min=1e-8)
gamma = float(config.get("snr_gamma", 1.0))
snr_weight = snr / (snr + gamma)
loss_cont = (loss_cont * snr_weight.mean()).mean()
loss_disc = 0.0 loss_disc = 0.0
loss_disc_count = 0 loss_disc_count = 0
for i, logit in enumerate(logits): for i, logit in enumerate(logits):
@@ -360,6 +370,22 @@ def main():
q_gen = torch.quantile(x_gen, q_tensor, dim=0) q_gen = torch.quantile(x_gen, q_tensor, dim=0)
quantile_loss = torch.mean(torch.abs(q_gen - q_real)) quantile_loss = torch.mean(torch.abs(q_gen - q_real))
loss = loss + q_weight * quantile_loss loss = loss + q_weight * quantile_loss
stat_weight = float(config.get("residual_stat_weight", 0.0))
if stat_weight > 0:
# residual distribution matching (mean/std)
a_bar_t = alphas_cumprod[t].view(-1, 1, 1)
if cont_target == "x0":
x_gen = eps_pred
else:
x_gen = (x_cont_t - torch.sqrt(1.0 - a_bar_t) * eps_pred) / torch.sqrt(a_bar_t)
x_real = x_cont_resid
mean_real = x_real.mean(dim=(0, 1))
mean_gen = x_gen.mean(dim=(0, 1))
std_real = x_real.std(dim=(0, 1))
std_gen = x_gen.std(dim=(0, 1))
stat_loss = F.mse_loss(mean_gen, mean_real) + F.mse_loss(std_gen, std_real)
loss = loss + stat_weight * stat_loss
opt.zero_grad() opt.zero_grad()
loss.backward() loss.backward()
if float(config.get("grad_clip", 0.0)) > 0: if float(config.get("grad_clip", 0.0)) > 0: