search

NVlabs / AL-SSL

Other
18 stars 5 forks source link

Data Loader error #1

Open erivera1802 opened 1 year ago

erivera1802 commented 1 year ago

Hi, Im trying to run the code with the COCO dataset, but I get an error when trying to extract the images from the data loader, specifically:

` Loading the dataset... Training SSD on: MS COCO Using the specified args: Loading the backbone pretrained in Imagenet... 0 Traceback (most recent call last): File "/home/esteban/Promotion/ComputerVision/AL-SSL/train.py", line 255, in train images, targets, semis = next(batch_iterator) File "/home/esteban/anaconda3/envs/ki/lib/python3.10/site-packages/torch/utils/data/dataloader.py", line 634, in next data = self._next_data() File "/home/esteban/anaconda3/envs/ki/lib/python3.10/site-packages/torch/utils/data/dataloader.py", line 1326, in _next_data return self._process_data(data) File "/home/esteban/anaconda3/envs/ki/lib/python3.10/site-packages/torch/utils/data/dataloader.py", line 1372, in _process_data data.reraise() File "/home/esteban/anaconda3/envs/ki/lib/python3.10/site-packages/torch/_utils.py", line 644, in reraise raise exception ValueError: Caught ValueError in DataLoader worker process 1. Original Traceback (most recent call last): File "/home/esteban/anaconda3/envs/ki/lib/python3.10/site-packages/torch/utils/data/_utils/worker.py", line 308, in _worker_loop data = fetcher.fetch(index) File "/home/esteban/anaconda3/envs/ki/lib/python3.10/site-packages/torch/utils/data/_utils/fetch.py", line 51, in fetch data = [self.dataset[idx] for idx in possibly_batched_index] File "/home/esteban/anaconda3/envs/ki/lib/python3.10/site-packages/torch/utils/data/_utils/fetch.py", line 51, in data = [self.dataset[idx] for idx in possibly_batched_index] File "/home/esteban/Promotion/ComputerVision/AL-SSL/data/coco.py", line 132, in getitem im, gt, h, w, semi = self.pull_item(index) File "/home/esteban/Promotion/ComputerVision/AL-SSL/data/coco.py", line 171, in pull_item img, boxes, labels = self.transform(img, target[:, :4], File "/home/esteban/Promotion/ComputerVision/AL-SSL/utils/augmentations.py", line 425, in call return self.augment(img, boxes, labels) File "/home/esteban/Promotion/ComputerVision/AL-SSL/utils/augmentations.py", line 52, in call img, boxes, labels = t(img, boxes, labels) File "/home/esteban/Promotion/ComputerVision/AL-SSL/utils/augmentations.py", line 238, in call mode = random.choice(self.sample_options) File "mtrand.pyx", line 920, in numpy.random.mtrand.RandomState.choice ValueError: setting an array element with a sequence. The requested array has an inhomogeneous shape after 1 dimensions. The detected shape was (6,) + inhomogeneous part.

`

I have only changed the paths of the datasets, but somehow this error appears, do you know what It could be?

bharanibala commented 6 months ago

Hi,

I am also facing the same issue but I am using a custom dataset. Did you solve this issue? If so, Could you please help me?

Thanks, Bharaneeshwar.