FileNotFoundError: [Errno 2] No such file or directory: '/cluster/balrog/jtang/3D-FUTURE-model/2924a537-18e9-428d-9922-ab1460e1f58b/raw_model_norm_pc_lat32.npz'
Traceback (most recent call last):
File "generate_diffusion.py", line 472, in
How to solve it, how can I change the location of "/cluster/balrog/jtang/"
I think that location is written in the .pkl file(preprocessed dataset) and it can't be changed,so you may have to do you own preprosessing
Same problem. Have you solved it? @StephenChen2016
i feel it's a data path problem. you need to change to the data path in your own machine.
I think that location is written in the .pkl file(preprocessed dataset) and it can't be changed,so you may have to do you own preprosessing
if you download the preprocessed dataset from my gdrive link, then you can skip the data preprocessing. But you still need to change the data path in your own machine.
I think that location is written in the .pkl file(preprocessed dataset) and it can't be changed,so you may have to do you own preprosessing
if you download the preprocessed dataset from my gdrive link, then you can skip the data preprocessing. But you still need to change the data path in your own machine.
i have checked all the place to change the data path, there is not any other place for users to change.I think maybe he is right
if you modify the path_to_models forcibly in class ThreedFutureModel from /DiffuScene/scene_synthesis/datasets/threed_front_scene.py, the file generate.sh can be run properly
The "/cluster/balrog/jtang/" seem to be written in the .pkl file, which you would not like to modify.
ThreedFutureDataset.from_pickled_dataset creates a dataset directly without __init__, so modifying ThreedFutureModel.__init__ won't work either.
If you don't want to do the pre-processing work, it may be helpful to iteratively set path_to_models to your directory of 3D-FUTURE-DATASET in scene_synthesis/datasets/threed_future_dataset.py: from_pickled_dataset after pickle.load like
@classmethod
def from_pickled_dataset(cls, path_to_pickled_dataset):
with open(path_to_pickled_dataset, "rb") as f:
dataset = pickle.load(f)
for obj in dataset:
obj.path_to_models = "some 3D-FUTURE dir" # TODO - set to correct path
return dataset
@huqng Thank you!