Question about weights

[mah-asd]

Hello, Thanks for your great job sharing with us, could you please help me with getting access to the model weights during the training?

I see in rendering there is weight but it's empty as the function returns [] as weights:

rgb_map, alphas_map, depth_map, weights, uncertainty = renderer(rays_train, tensorf, chunk=args.batch_size,
N_samples=nSamples, white_bg = white_bg, ndc_ray=ndc_ray, device=device, is_train=True)

I figured out with this function:

def get_param(tensorf, logfolder, expname, model_name):
tensorf.save(f'{logfolder}/{expname}.th')
ckpt = torch.load(f'{logfolder}/{expname}.th', map_location=device)
param = []
for key in ckpt['state_dict'].keys():
param.append(ckpt['state_dict'][str(key)].reshape(-1).clone().detach())
params = torch.cat(param)
return params

Although it's writing the model, then loading it, then concatenate all the matrices, am I missing something?