patch_ Loader cannot iterate.

Is there an existing issue for this?

• I have searched the existing issues

Bug summary

I hope that the fixed size patch size will be used in model reasoning, and the data will be generated into subjects according to the tutorial.After the dataloader loads the patch, I want to iterate over the data through the for loop for reasoning. But I’ve been patching_ The loader iteration reports an error. The error message is: raise filenotfounderror (f’file not found: “{path}” ') FileNotFoundError: File not found: “affine”

Code for reproduction

class trochio_test_dataset(Dataset):
    def __init__(self,data_list):
        self.test_images=data_list
        print(self.test_images)
        self.patch_size=(80,160,160)
        self.volume=4,4,4
        self.subject=[]
        for i , image_path in enumerate(list(self.test_images)):
            print("data_process"+str(i))
            print(image_path)
            subject = tio.Subject(source=tio.data.ScalarImage(image_path))
            self.subject.append(subject)
        self.subject_dataset=tio.SubjectsDataset(self.subject)
def predict(opt):
    #define input type
    type=0
    #paths or files
    if os.path.isfile(opt.data[0]):
        pre_input=opt.data
        type=1
    elif os.path.isdir(opt.data[0]):
        pre_input=opt.data
        type=2
    else:
        logging.info("please check up your input,just accept files (.nii .dicom .mhd) or folders ")
        sys.exit(0)
    print(type)
    #load model
    model = Trans_Unet(c1=1, c2=8)
    device=torch.device('cuda' if torch.cuda.is_available() else 'cpu')
    # logging.info(f"Loading model {opt.weights}")
    logging.info(f"Using device {device}")
    model.to(device=device)

    logging.info(f"Model load!")
    """
    #load weight
    if torch.cuda.is_available():
        model=torch.load(opt.weights,map_location=torch.device("cuda"))
    else:
        model=torch.load(opt.weights,map_location=torch.device("cpu"))
    model.load_state_dict(model)
    """
    #set eval
    model.eval()
    #load file
    if type==1:
        input_files=pre_input
    elif type==2:
        input_files=[]
        for f in range(len(pre_input)):
            print(f"input folders as flow:{pre_input[f]}")
            input_file=subfiles(pre_input[f],suffix='nii.gz')#[pre_input[f]]
            input_files.extend(input_file)
    #create subject dataset
    subject_dataset=trochio_test_dataset(input_files).subject_dataset
    print(f"subject dataset {subject_dataset}")
    #iter subject
    for i in range(len(subject_dataset)):
        grid_sampler = tio.inference.GridSampler(
            subject_dataset[i],  # some NumPy array
            patch_size=(80, 160, 160),
            patch_overlap=4,
        )
        patch_loader = torch.utils.data.DataLoader(grid_sampler, batch_size=8)
        print(grid_sampler.subject)
        aggregator = tio.inference.GridAggregator(grid_sampler)
        print(aggregator.patch_overlap)
        with torch.no_grad():
            for patches_batch in patch_loader:
                print(1111111111)
                input_tensor = patches_batch['source'][tio.DATA]
                #test  --not enter model
                print(input_tensor)
                locations = patches_batch[tio.LOCATION]
                logging.info("OK")
                aggregator.add_batch(input_tensor, locations)
        output_tensor = aggregator.get_output_tensor()

Actual outcome

FileNotFoundError: File not found: “affine”:

Error messages

Traceback (most recent call last):
  File "D:/xxxx/MODELS/Li_Project/test.py", line 127, in <module>
    main(opt)
  File "D:/xxxx/MODELS/Li_Project/test.py", line 117, in main
    predict(opt)
  File "D:/xxxx/MODELS/Li_Project/test.py", line 82, in predict
    for patches_batch in patch_loader:
  File "D:\Anaconda3\install\envs\conda\lib\site-packages\torch\utils\data\dataloader.py", line 530, in __next__
    data = self._next_data()
  File "D:\Anaconda3\install\envs\conda\lib\site-packages\torch\utils\data\dataloader.py", line 570, in _next_data
    data = self._dataset_fetcher.fetch(index)  # may raise StopIteration
  File "D:\Anaconda3\install\envs\conda\lib\site-packages\torch\utils\data\_utils\fetch.py", line 52, in fetch
    return self.collate_fn(data)
  File "D:\Anaconda3\install\envs\conda\lib\site-packages\torch\utils\data\_utils\collate.py", line 157, in default_collate
    return elem_type({key: default_collate([d[key] for d in batch]) for key in elem})
  File "D:\Anaconda3\install\envs\conda\lib\site-packages\torch\utils\data\_utils\collate.py", line 157, in <dictcomp>
    return elem_type({key: default_collate([d[key] for d in batch]) for key in elem})
  File "D:\Anaconda3\install\envs\conda\lib\site-packages\torch\utils\data\_utils\collate.py", line 157, in default_collate
    return elem_type({key: default_collate([d[key] for d in batch]) for key in elem})
  File "D:\Anaconda3\install\envs\conda\lib\site-packages\torchio\data\image.py", line 746, in __init__
    super().__init__(*args, **kwargs)
  File "D:\Anaconda3\install\envs\conda\lib\site-packages\torchio\data\image.py", line 161, in __init__
    self.path = self._parse_path(path)
  File "D:\Anaconda3\install\envs\conda\lib\site-packages\torchio\data\image.py", line 428, in _parse_path
    return [self._parse_single_path(p) for p in path]
  File "D:\Anaconda3\install\envs\conda\lib\site-packages\torchio\data\image.py", line 428, in <listcomp>
    return [self._parse_single_path(p) for p in path]
  File "D:\Anaconda3\install\envs\conda\lib\site-packages\torchio\data\image.py", line 418, in _parse_single_path
    raise FileNotFoundError(f'File not found: "{path}"')
FileNotFoundError: File not found: "affine":

Expected outcome

I want to be able to use patch inference

System info

No response

[LucaLumetti]

Hello difficult to reproduce, and the File not found error should a path problem on your side … Can you reproduce it with a minimal example involving torchio dataset ?

hello,please test this example. import torch import torch.nn as nn import torchio as tio patch_overlap = 4, 4, 4 # or just 4 patch_size = 88, 88, 60 subject = tio.datasets.Colin27() subject grid_sampler = tio.inference.GridSampler( subject, patch_size, patch_overlap, ) patch_loader = torch.utils.data.DataLoader(grid_sampler, batch_size=4) aggregator = tio.inference.GridAggregator(grid_sampler) model = nn.Identity().eval() with torch.no_grad(): for patches_batch in patch_loader: input_tensor = patches_batch['t1'][tio.DATA] locations = patches_batch[tio.LOCATION] logits = model(input_tensor) labels = logits.argmax(dim=tio.CHANNELS_DIMENSION, keepdim=True) outputs = labels aggregator.add_batch(outputs, locations) output_tensor = aggregator.get_output_tensor()

I cannot reproduce this bug too, even with your code. Can you paste the error message that this code outputs to you?

[fepegar]

Hi, @boundaryT. It is hard for us to help you if you don’t provide a minimal reproducible example, as requested in the issue template.

I don’t know what opt or subfiles are in your code, but they might be generating wrong filenames. If you are able to provide a example we can reproduce, feel free to open a new issue.