ChenFengYe / motion-latent-diffusion

[CVPR 2023] Executing your Commands via Motion Diffusion in Latent Space, a fast and high-quality motion diffusion model
https://chenxin.tech/mld/
MIT License
586 stars 55 forks source link

Error with "name_list" and "length_list" in datasets #53

Open alsokh opened 1 year ago

alsokh commented 1 year ago

Congrats on your great paper!

I am running the "python demo.py --cfg ./configs/config_mld_humanml3d.yaml --cfg_assets ./configs/assets.yaml --example ./demo/example.txt" to get demo results on my colab. First of all, data should be manually downloaded from HumanML3D, because I got the error, stated that Mean.npy and Std.npy doesn't exist. After struggling with that and resolving it with hard code now I got the following error:

!python demo.py --cfg ./configs/config_mld_humanml3d.yaml --cfg_assets ./configs/assets.yaml --example ./demo/example.txt

2023-09-08 07:55:01.073355: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT Length-50: a man kicks with something or someone with his left leg. Length-100: A person is skipping rope. Length-100: a person walks backward slowly. Traceback (most recent call last): File "/content/demo.py", line 337, in main() File "/content/demo.py", line 101, in main dataset = get_datasets(cfg, logger=logger, phase="test")[0] File "/content/mld/data/get_data.py", line 92, in get_datasets dataset = dataset_module_map[dataset_name.lower()]( File "/content/mld/data/HumanML3D.py", line 36, in init self._sample_set = self.get_sample_set(overrides=sample_overrides) File "/content/mld/data/base.py", line 37, in get_sample_set return self.Dataset(split_file=split_file, **sample_params) File "/content/mld/data/humanml/data/dataset.py", line 357, in init name_list, lengthlist = zip( ValueError: not enough values to unpack (expected 2, got 0)

Also, when I manually ignore this part, I get an error stating that motion is called before being assigned.

Thank you in advance for your advice.

jercas commented 1 year ago

hi! I have also encountered the same problem. Have you solved this problem?

santurini commented 11 months ago

Should be solved by unzipping texts.zip in ./datasets/humanml3d

yyvhang commented 11 months ago

Should be solved by unzipping texts.zip in ./datasets/humanml3d

still does not work, this error also exists

santurini commented 11 months ago

still does not work, this error also exists

Can you share how it is structured your folder?

yyvhang commented 11 months ago

still does not work, this error also exists

Can you share how it is structured your folder?

I have solved it, thks!

zhangjingze21 commented 11 months ago

I still have the problem after I unzipped tests.zip in ./datasets/humanml3d, however I still face the problem, could you please tell me how did you solve the problem? Thank you!

mkstmyk commented 7 months ago

For me, it was solved by downloading KIT_ML, unrar all rar files in it, and update configs/assets.yaml to specify where it is placed as like follows:

  HUMANML3D:
    ROOT: '../HumanML3D/KIT-ML/' # HumanML3D directory
    SPLIT_ROOT: '../HumanML3D/KIT-ML/' # HumanML3D splits directory

UPDATE: By using KIT_ML, you may face the known problem: I had to download and preprocess correct data from HumanML3D Hope this helps.