Guoxni
commented
4 days ago Run process.py again
At 2024-11-25 21:34:27, "weida wang" @.***> wrote:
I met the same issue, did you solve it ?
— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>
root@autodl-container-47f511bde8-27ae324d:~/autodl-tmp/GaussianTalker-main# python train.py -s data/May --model_path output --configs arguments/64_dim_1_transformer.py Optimizing output Output folder: output [19/11 00:20:33] feature_dim: 64 [19/11 00:20:34] [INFO] load aud_features: torch.Size([6073, 29, 16]) [19/11 00:20:34] Reading Training Transforms [19/11 00:20:34] Reading Test Transforms [19/11 00:20:42] Generating Video Transforms [19/11 00:20:46] Loading Training Cameras [19/11 00:20:48] Loading Test Cameras [19/11 00:20:48] Loading Video Cameras [19/11 00:20:48] Deformation Net Set aabb [0.78709733 0.8721499 0.45940906] [-0.7947565 -0.8129441 -0.55458224] [19/11 00:20:48] Voxel Plane: set aabb= Parameter containing: tensor([[ 0.7871, 0.8721, 0.4594], [-0.7948, -0.8129, -0.5546]]) [19/11 00:20:48] Number of points at initialisation : 34650 [19/11 00:20:49] ['xyz', 'deformation', 'grid', 'f_dc', 'f_rest', 'opacity', 'scaling', 'rotation'] [19/11 00:20:49] Training progress: 0%| | 0/7999 [00:00<?, ?it/s]data loading done [19/11 00:20:49] Traceback (most recent call last): File "train.py", line 408, in
training(lp.extract(args), hp.extract(args), op.extract(args), pp.extract(args), args.test_iterations, args.save_iterations, args.checkpoint_iterations, args.start_checkpoint, args.debug_from, args.expname, args.use_wandb)
File "train.py", line 278, in training
gaussians, scene, "coarse", tb_writer, opt.coarse_iterations,timer, use_wandb)
File "train.py", line 152, in scene_reconstruction
output = render_from_batch(viewpoint_cams, gaussians, pipe, random_color, stage=stage, batch_size=batch_size, canonical_tri_plane_factor_list=opt.canonical_tri_plane_factor_list,iteration=iteration)
File "/root/autodl-tmp/GaussianTalker-main/gaussian_renderer/init.py", line 162, in render_from_batch
cov3D_precomp = cov3D_precomp,)
ValueError: too many values to unpack (expected 3)
Training progress: 0%| | 0/7999 [00:06<?, ?it/s]
Has anyone encountered such a problem before?