A high-performance image processing library designed to optimize and extend the Albumentations library with specialized functions for advanced image transformations. Perfect for developers working in computer vision who require efficient and scalable image augmentation.
MIT License
12
stars
3
forks
source link
YOLOv8 with Albucore >= 0.0.15 results in masks must be list of numpy arrays #35
I've been using Albumentations in YOLOv8 for instance segmentation.
Ultralytics versions have no problem, but albumentations results into a conflict.
The error starts after Albumentations version > 1.4.7
Versions >=1.4.8 and <=1.4.15 don't get recognized by ultralytics at all (starting training process, the albumentations init method for data augmentation in ultralytics.data.augment doesn't start, which as far as I know mostly means that ultralytics doesn't see albumentations installed)
Versions >=1.4.16 get recognized again, but result into the following error:
which made 1.4.14 recognizable again and no numpy array error.
Would be great if whatever happend after albucre 0.0.14 can be fixed.
Here is the full error code:
TypeError Traceback (most recent call last)
Cell In[6], line 25
23 initYaml(i)
24 model = YOLO('yolov8x-seg.pt')
---> 25 model_results = model.train(data=os.path.join(ROOT_DIR, f'config{i}.yaml'), epochs=training_epochs, imgsz=512, batch=16, deterministic=True, plots=True,
26 close_mosaic = 0,
27 # disable built-in augmentation, instead use Albumentations Library
28 augment=False, hsv_h=0, hsv_s=0, hsv_v=0, degrees=0, translate=0,
29 scale=0, shear=0.0, perspective=0, flipud=0, fliplr=0, bgr=0,
30 mosaic=0, mixup=0, copy_paste=0, erasing=0, crop_fraction=0)
31 results = model.val()
33 print("\n" + "#" * 60)
File [~/yolov8/lib/python3.11/site-packages/ultralytics/engine/model.py:802](http://172.22.50.111:9000/user/user/lab/tree/myraid/YOLOv8_home/~/yolov8/lib/python3.11/site-packages/ultralytics/engine/model.py#line=801), in Model.train(self, trainer, **kwargs)
799 self.model = self.trainer.model
801 self.trainer.hub_session = self.session # attach optional HUB session
--> 802 self.trainer.train()
803 # Update model and cfg after training
804 if RANK in {-1, 0}:
File [~/yolov8/lib/python3.11/site-packages/ultralytics/engine/trainer.py:207](http://172.22.50.111:9000/user/user/lab/tree/myraid/YOLOv8_home/~/yolov8/lib/python3.11/site-packages/ultralytics/engine/trainer.py#line=206), in BaseTrainer.train(self)
204 ddp_cleanup(self, str(file))
206 else:
--> 207 self._do_train(world_size)
File [~/yolov8/lib/python3.11/site-packages/ultralytics/engine/trainer.py:367](http://172.22.50.111:9000/user/user/lab/tree/myraid/YOLOv8_home/~/yolov8/lib/python3.11/site-packages/ultralytics/engine/trainer.py#line=366), in BaseTrainer._do_train(self, world_size)
365 pbar = TQDM(enumerate(self.train_loader), total=nb)
366 self.tloss = None
--> 367 for i, batch in pbar:
368 self.run_callbacks("on_train_batch_start")
369 # Warmup
File [~/yolov8/lib/python3.11/site-packages/tqdm/std.py:1181](http://172.22.50.111:9000/user/user/lab/tree/myraid/YOLOv8_home/~/yolov8/lib/python3.11/site-packages/tqdm/std.py#line=1180), in tqdm.__iter__(self)
1178 time = self._time
1180 try:
-> 1181 for obj in iterable:
1182 yield obj
1183 # Update and possibly print the progressbar.
1184 # Note: does not call self.update(1) for speed optimisation.
File [~/yolov8/lib/python3.11/site-packages/ultralytics/data/build.py:48](http://172.22.50.111:9000/user/user/lab/tree/myraid/YOLOv8_home/~/yolov8/lib/python3.11/site-packages/ultralytics/data/build.py#line=47), in InfiniteDataLoader.__iter__(self)
46 """Creates a sampler that repeats indefinitely."""
47 for _ in range(len(self)):
---> 48 yield next(self.iterator)
File [~/yolov8/lib/python3.11/site-packages/torch/utils/data/dataloader.py:630](http://172.22.50.111:9000/user/user/lab/tree/myraid/YOLOv8_home/~/yolov8/lib/python3.11/site-packages/torch/utils/data/dataloader.py#line=629), in _BaseDataLoaderIter.__next__(self)
627 if self._sampler_iter is None:
628 # TODO(https://github.com/pytorch/pytorch/issues/76750)
629 self._reset() # type: ignore[call-arg]
--> 630 data = self._next_data()
631 self._num_yielded += 1
632 if self._dataset_kind == _DatasetKind.Iterable and \
633 self._IterableDataset_len_called is not None and \
634 self._num_yielded > self._IterableDataset_len_called:
File ~/yolov8/lib/python3.11/site-packages/torch/utils/data/dataloader.py:1344, in _MultiProcessingDataLoaderIter._next_data(self)
1342 else:
1343 del self._task_info[idx]
-> 1344 return self._process_data(data)
File [~/yolov8/lib/python3.11/site-packages/torch/utils/data/dataloader.py:1370](http://172.22.50.111:9000/user/user/lab/tree/myraid/YOLOv8_home/~/yolov8/lib/python3.11/site-packages/torch/utils/data/dataloader.py#line=1369), in _MultiProcessingDataLoaderIter._process_data(self, data)
1368 self._try_put_index()
1369 if isinstance(data, ExceptionWrapper):
-> 1370 data.reraise()
1371 return data
File [~/yolov8/lib/python3.11/site-packages/torch/_utils.py:706](http://172.22.50.111:9000/user/user/lab/tree/myraid/YOLOv8_home/~/yolov8/lib/python3.11/site-packages/torch/_utils.py#line=705), in ExceptionWrapper.reraise(self)
702 except TypeError:
703 # If the exception takes multiple arguments, don't try to
704 # instantiate since we don't know how to
705 raise RuntimeError(msg) from None
--> 706 raise exception
TypeError: Caught TypeError in DataLoader worker process 0.
Original Traceback (most recent call last):
File "/home/user/yolov8/lib/python3.11/site-packages/torch/utils/data/_utils/worker.py", line 309, in _worker_loop
data = fetcher.fetch(index) # type: ignore[possibly-undefined]
^^^^^^^^^^^^^^^^^^^^
File "/home/user/yolov8/lib/python3.11/site-packages/torch/utils/data/_utils/fetch.py", line 52, in fetch
data = [self.dataset[idx] for idx in possibly_batched_index]
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/user/yolov8/lib/python3.11/site-packages/torch/utils/data/_utils/fetch.py", line 52, in <listcomp>
data = [self.dataset[idx] for idx in possibly_batched_index]
~~~~~~~~~~~~^^^^^
File "/home/user/yolov8/lib/python3.11/site-packages/ultralytics/data/base.py", line 256, in __getitem__
return self.transforms(self.get_image_and_label(index))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/user/yolov8/lib/python3.11/site-packages/ultralytics/data/augment.py", line 201, in __call__
data = t(data)
^^^^^^^
File "/tmp/ipykernel_39409/4044539362.py", line 189, in __call__
new = self.transform(image=im,
^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/user/yolov8/lib/python3.11/site-packages/albumentations/core/composition.py", line 334, in __call__
self.preprocess(data)
File "/home/user/yolov8/lib/python3.11/site-packages/albumentations/core/composition.py", line 363, in preprocess
self._check_args(**data)
File "/home/user/yolov8/lib/python3.11/site-packages/albumentations/core/composition.py", line 416, in _check_args
raise TypeError(f"{data_name} must be list of numpy arrays")
TypeError: masks must be list of numpy arrays
ternaus
commented
1 month ago
I've been using Albumentations in YOLOv8 for instance segmentation. Ultralytics versions have no problem, but albumentations results into a conflict.
The error starts after Albumentations version > 1.4.7
Versions >=1.4.8 and <=1.4.15 don't get recognized by ultralytics at all (starting training process, the albumentations init method for data augmentation in ultralytics.data.augment doesn't start, which as far as I know mostly means that ultralytics doesn't see albumentations installed)
Versions >=1.4.16 get recognized again, but result into the following error:
the only fix that worked mentioned here
https://github.com/ultralytics/ultralytics/pull/6624
was to downgrade albumentations and its core
which made 1.4.14 recognizable again and no numpy array error.
Would be great if whatever happend after albucre 0.0.14 can be fixed.
Here is the full error code: