Project-MONAI / Project-MONAI/MONAI

RandSpatialCropSamplesd

Offen
#7,828 2 Kommentare 0 Reaktionen 0 zugewiesene Personen Auf GitHub ansehen

Dieses Issue hat noch niemand übernommen.

Vorherrschende Sprache
Python
Sterne
8.7k
Forks
1.6k
Ø Merge
5 T. 1 Std.
Gemergte PRs (30 T.)
20

Beschreibung

Hi，I want to use batch_size = 2 for training. When using CenterSpatialCropd(keys=["image", "label"],roi_size=(224, 224, 224)), I can train the network, but when I use RandSpatialCropSamplesd(keys=["image", "label"],roi_size=(224, 224, 224), random_size=False, num_samples=2),ResizeWithPadOrCropd( keys=["image", "label"], spatial_size=224, mode='constant', ), the following error occurred.

collate/stack a list of tensors
E: stack expects each tensor to be equal size, but got [1, 275, 386, 386] at entry 0 and [1, 275, 380, 380] at entry 2, shape [torch.Size([1, 275, 386, 386]), torch.Size([1, 275, 386, 386]), torch.Size([1, 275, 380, 380]), torch.Size([1, 275, 380, 380])] in collate([tensor([[[[0., 0., 0., ..., 0., 0., 0.],
[0., 0., 0., ..., 0., 0., 0.],
[0., 0., 0., ..., 0., 0., 0.],
...,
[0., 0., 0., ..., 0., 0., 0.],
[0., 0., 0., ..., 0., 0., 0.],
[0., 0., 0., ..., 0., 0., 0.]],

     [[0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      ...,
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.]],

     [[0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      ...,
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.]],

     ...,

     [[0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      ...,
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.]],

     [[0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      ...,
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.]],

     [[0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      ...,
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.]]]]), tensor([[[[0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      ...,
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.]],

     [[0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      ...,
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.]],

     [[0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      ...,
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.]],

     ...,

     [[0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      ...,
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.]],

     [[0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      ...,
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.]],

     [[0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      ...,
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.]]]]), tensor([[[[0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      ...,
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.]],

     [[0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      ...,
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.]],

     [[0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      ...,
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.]],

     ...,

     [[0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      ...,
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.]],

     [[0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      ...,
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.]],

     [[0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      ...,
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.]]]]), tensor([[[[0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      ...,
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.]],

     [[0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      ...,
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.]],

     [[0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      ...,
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.]],

     ...,

     [[0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      ...,
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.]],

     [[0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      ...,
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.]],

     [[0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      ...,
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.],
      [0., 0., 0.,  ..., 0., 0., 0.]]]])])

collate dict key "image_meta_dict" out of 4 keys

collate dict key "sizeof_hdr" out of 43 keys

collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "extents" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "session_error" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "dim_info" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "dim" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "intent_p1" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "intent_p2" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "intent_p3" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "intent_code" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "datatype" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "bitpix" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "slice_start" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "pixdim" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "vox_offset" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "scl_slope" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "scl_inter" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "slice_end" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "slice_code" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "xyzt_units" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "cal_max" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "cal_min" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "slice_duration" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "toffset" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "glmax" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "glmin" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "qform_code" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "sform_code" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "quatern_b" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "quatern_c" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "quatern_d" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "qoffset_x" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "qoffset_y" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "qoffset_z" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "srow_x" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "srow_y" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "srow_z" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "affine" out of 43 keys
collate/stack a list of tensors
collate dict key "original_affine" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "as_closest_canonical" out of 43 keys
collate dict key "spatial_shape" out of 43 keys
collate/stack a list of numpy arrays
collate/stack a list of tensors
collate dict key "space" out of 43 keys
collate dict key "original_channel_dim" out of 43 keys
collate dict key "filename_or_obj" out of 43 keys
Traceback (most recent call last):
File "/mnt/store/zhengsch/Ostia_landmark/tools/train.py", line 572, in
Trainer().run()
File "/mnt/store/zhengsch/Ostia_landmark/tools/train.py", line 332, in run
self.train_step(epoch)
File "/mnt/store/zhengsch/Ostia_landmark/tools/train.py", line 392, in train_step
for i, batch_data in enumerate(self.trn_dl):
File "/opt/miniforge3/envs/monai/lib/python3.9/site-packages/torch/utils/data/dataloader.py", line 681, in next
data = self._next_data()
File "/opt/miniforge3/envs/monai/lib/python3.9/site-packages/torch/utils/data/dataloader.py", line 1376, in _next_data
return self._process_data(data)
File "/opt/miniforge3/envs/monai/lib/python3.9/site-packages/torch/utils/data/dataloader.py", line 1402, in _process_data
data.reraise()
File "/opt/miniforge3/envs/monai/lib/python3.9/site-packages/torch/_utils.py", line 461, in reraise
raise exception
RuntimeError: Caught RuntimeError in DataLoader worker process 0.
Original Traceback (most recent call last):
File "/home/zhengsch/.local/lib/python3.9/site-packages/monai/data/utils.py", line 516, in list_data_collate
ret = collate_fn(data)
File "/home/zhengsch/.local/lib/python3.9/site-packages/monai/data/utils.py", line 478, in collate_meta_tensor
return {k: collate_meta_tensor([d[k] for d in batch]) for k in elem_0}
File "/home/zhengsch/.local/lib/python3.9/site-packages/monai/data/utils.py", line 478, in
return {k: collate_meta_tensor([d[k] for d in batch]) for k in elem_0}
File "/home/zhengsch/.local/lib/python3.9/site-packages/monai/data/utils.py", line 483, in collate_meta_tensor
return default_collate(batch)
File "/opt/miniforge3/envs/monai/lib/python3.9/site-packages/torch/utils/data/utils/collate.py", line 140, in default_collate
out = elem.new(storage).resize(len(batch), *list(elem.size()))
RuntimeError: Trying to resize storage that is not resizable

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File "/opt/miniforge3/envs/monai/lib/python3.9/site-packages/torch/utils/data/_utils/worker.py", line 302, in _worker_loop
data = fetcher.fetch(index)
File "/opt/miniforge3/envs/monai/lib/python3.9/site-packages/torch/utils/data/_utils/fetch.py", line 52, in fetch
return self.collate_fn(data)
File "/home/zhengsch/.local/lib/python3.9/site-packages/monai/data/utils.py", line 696, in pad_list_data_collate
return PadListDataCollate(method=method, mode=mode, **kwargs)(batch)
File "/home/zhengsch/.local/lib/python3.9/site-packages/monai/transforms/croppad/batch.py", line 114, in call
return list_data_collate(batch)
File "/home/zhengsch/.local/lib/python3.9/site-packages/monai/data/utils.py", line 529, in list_data_collate
raise RuntimeError(re_str) from re
RuntimeError: Trying to resize storage that is not resizable

Beitragsleitfaden

Beitragsleitfaden öffnen

Erste Schritte

  1. Lies das ganze Issue und danach den Beitragsleitfaden des Projekts.
  2. Schreib ins Issue, dass du es übernimmst — das erspart doppelte Arbeit.
  3. Forke das Repository und arbeite in einem Branch.
  4. Öffne einen Pull Request, der die Issue-Nummer nennt.

Rechercherichtung

Reproduzieren Sie die gemeldete Pipeline mit RandSpatialCropSamplesd, ResizeWithPadOrCropd und batch_size=2. Untersuchen Sie anschließend den Collate-Fehler und die unterschiedlichen Tensorformen. Ermitteln Sie das erwartete Verhalten der gesampelten Ausgaben und verifizieren Sie, dass die gemeldete Batch-Konfiguration entweder Tensoren mit konsistenten Formen erzeugt oder eindeutig als nicht unterstützt dokumentiert ist.

Vom Indexierungsmodell aus dem Issue-Text verfasst.

Bewertung

Tech-Stack
python, pytorch
Bereich
data, machine-learning
Issue-Typ
Bug
Schwierigkeit
4/5
Geschätzter Aufwand
3-5 Tage
Aktivitätsstatus
Veraltet
Klarheit
Muss geklärt werden
Anfängerfreundlichkeit
25/100

Neue Issues direkt in Ihr Postfach

Eine kurze Übersicht über anfängerfreundliche GitHub-Issues.