search

hugohadfield / pytorch_image_regession

A template for doing regression from images with pytorch
MIT License
17 stars 2 forks source link

Dataloader expects classes #1

Open garygazanga opened 1 month ago

garygazanga commented 1 month ago

I tried to run your data_loader.py on some images organised to suit your code (i.e. topDir/train/images.png and topDir/test/images.png and topDir/train.csv and test.csv with each line as filename, val1, val2). However, I get the following error: Traceback (most recent call last): File "dataLoader.py", line 159, in data = RegressionTaskData() File "dataLoader.py", line 105, in init self.trainloader = self.make_trainloader() File "dataLoader.py", line 120, in make_trainloader train_data = RegressionImageFolder( File "dataLoader.py", line 84, in init super().init(root, *args, **kwargs) File "/usr/local/lib/python3.8/dist-packages/torchvision/datasets/folder.py", line 309, in init super().init( File "/usr/local/lib/python3.8/dist-packages/torchvision/datasets/folder.py", line 144, in init classes, class_to_idx = self.find_classes(self.root) File "/usr/local/lib/python3.8/dist-packages/torchvision/datasets/folder.py", line 218, in find_classes return find_classes(directory) File "/usr/local/lib/python3.8/dist-packages/torchvision/datasets/folder.py", line 42, in find_classes raise FileNotFoundError(f"Couldn't find any class folder in {directory}.") FileNotFoundError: Couldn't find any class folder in /mnt/shared/Gary/motionTrain/train.

Why would it be looking for classes in a linear regression example?

garygazanga commented 1 month ago

It would appear that a single class subdir (of any name) in each of 'train' and 'test' gets around the problem.