search

NUS-HPC-AI-Lab / Neural-Network-Parameter-Diffusion

We introduce a novel approach for parameter generation, named neural network parameter diffusion (p-diff), which employs a standard latent diffusion model to synthesize a new set of parameters
787 stars 38 forks source link

i want to know why set len(param)==10 here? #21

Closed freja-zy closed 2 weeks ago

freja-zy commented 4 weeks ago

i want to know why set len(param)==10 here?

if i > = epoch: parameters.append(state_part(train_layer, net)) save_model_accs.append(acc) if len(parameters) == 10 or i == all_epoch - 1: torch.save(parameters, os.path.join(tmp_path, "pdata{}.pt".format(i))) parameters = []

1zeryu commented 3 weeks ago

To avoid storing too many models in one file, we set a maximum of 10 model parameters in one file. This is not a very important detail, you can also set this storage interval to 100