Error while running test.py

[kvuong2711]

Hi authors,

Thanks for the great work. I'm running the evaluation script for zeronvs-finetune and running into some error in terms of shape. Do you know what could be happening here:

Command:

python test.py -e configs/zeronvs_finetune -r 90000 -s zeronvs_evaluation -b 1 -w 4 --save_generations True

Traceback (most recent call last):
  File "test.py", line 178, in <module>
    main()
  File "test.py", line 113, in main
    pred = img_logger.log_img(model, batch, resume, split='test', returngrid='train', warpeddepth=None, onlyretimg=True).permute(0,2,3,1) # from chw to hwc, in range -1,1
  File "/home/kvuong/workspace/local/CVPR2025/nvs/ldm/logger.py", line 124, in log_img
    images = model.log_images(batch, no_recon_cond=self.no_recon_cond, only_sample_output=self.only_sample_output, split=split, **self.log_images_kwargs)
  File "/home/kvuong/miniconda3/envs/megascenes/lib/python3.8/site-packages/torch/autograd/grad_mode.py", line 27, in decorate_context
    return func(*args, **kwargs)
  File "/home/kvuong/workspace/local/CVPR2025/nvs/ldm/models/diffusion/ddpm.py", line 1592, in log_images
    samples_cfg, _ = self.sample_log(cond=c, batch_size=N, ddim=use_ddim,
  File "/home/kvuong/miniconda3/envs/megascenes/lib/python3.8/site-packages/torch/autograd/grad_mode.py", line 27, in decorate_context
    return func(*args, **kwargs)
  File "/home/kvuong/workspace/local/CVPR2025/nvs/ldm/models/diffusion/ddpm.py", line 1445, in sample_log
    outs = ddim_sampler.sample(ddim_steps, batch_size,
  File "/home/kvuong/miniconda3/envs/megascenes/lib/python3.8/site-packages/torch/autograd/grad_mode.py", line 27, in decorate_context
    return func(*args, **kwargs)
  File "/home/kvuong/workspace/local/CVPR2025/nvs/ldm/models/diffusion/ddim.py", line 109, in sample
    outs = self.ddim_sampling(conditioning, size,
  File "/home/kvuong/miniconda3/envs/megascenes/lib/python3.8/site-packages/torch/autograd/grad_mode.py", line 27, in decorate_context
    return func(*args, **kwargs)
  File "/home/kvuong/workspace/local/CVPR2025/nvs/ldm/models/diffusion/ddim.py", line 168, in ddim_sampling
    outs = self.p_sample_ddim(img, cond, ts, index=index, use_original_steps=ddim_use_original_steps,
  File "/home/kvuong/miniconda3/envs/megascenes/lib/python3.8/site-packages/torch/autograd/grad_mode.py", line 27, in decorate_context
    return func(*args, **kwargs)
  File "/home/kvuong/workspace/local/CVPR2025/nvs/ldm/models/diffusion/ddim.py", line 215, in p_sample_ddim
    c_in[k] = [torch.cat([ unconditional_conditioning[k][i], c[k][i]]) for i in range(len(c[k]))]
  File "/home/kvuong/workspace/local/CVPR2025/nvs/ldm/models/diffusion/ddim.py", line 215, in <listcomp>
    c_in[k] = [torch.cat([ unconditional_conditioning[k][i], c[k][i]]) for i in range(len(c[k]))]
RuntimeError: Sizes of tensors must match except in dimension 0. Expected size 225 but got size 168 for tensor number 1 in the list.