fasterrcnn-pytorch-training-pipeline

fasterrcnn-pytorch-training-pipeline copied to clipboard

I trained on my private dataset using the same format as VOC2007 but encountered this error. I can run on VOC2007 successfully

Traceback (most recent call last): File "/datashare3/charis/code/recaps/fasterRcnn/fasterrcnn-pytorch-training-pipeline-main/train.py", line 571, in main(args) File "/datashare3/charis/code/recaps/fasterRcnn/fasterrcnn-pytorch-training-pipeline-main/train.py", line 420, in main scaler=SCALER File "/datashare3/charis/code/recaps/fasterRcnn/fasterrcnn-pytorch-training-pipeline-main/torch_utils/engine.py", line 45, in train_one_epoch for images, targets in metric_logger.log_every(data_loader, print_freq, header): File "/datashare3/charis/code/recaps/fasterRcnn/fasterrcnn-pytorch-training-pipeline-main/torch_utils/utils.py", line 173, in log_every for obj in iterable: File "/datashare3/charis/anaconda/envs/simp/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 628, in next data = self._next_data() File "/datashare3/charis/anaconda/envs/simp/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 1333, in _next_data return self._process_data(data) File "/datashare3/charis/anaconda/envs/simp/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 1359, in _process_data data.reraise() File "/datashare3/charis/anaconda/envs/simp/lib/python3.7/site-packages/torch/_utils.py", line 543, in reraise raise exception ValueError: Caught ValueError in DataLoader worker process 0. Original Traceback (most recent call last): File "/datashare3/charis/anaconda/envs/simp/lib/python3.7/site-packages/torch/utils/data/_utils/worker.py", line 302, in _worker_loop data = fetcher.fetch(index) File "/datashare3/charis/anaconda/envs/simp/lib/python3.7/site-packages/torch/utils/data/_utils/fetch.py", line 58, in fetch data = [self.dataset[idx] for idx in possibly_batched_index] File "/datashare3/charis/anaconda/envs/simp/lib/python3.7/site-packages/torch/utils/data/_utils/fetch.py", line 58, in data = [self.dataset[idx] for idx in possibly_batched_index] File "/datashare3/charis/code/recaps/fasterRcnn/fasterrcnn-pytorch-training-pipeline-main/datasets.py", line 316, in getitem labels=labels) File "/datashare3/charis/anaconda/envs/simp/lib/python3.7/site-packages/albumentations/core/composition.py", line 207, in call p.preprocess(data) File "/datashare3/charis/anaconda/envs/simp/lib/python3.7/site-packages/albumentations/core/utils.py", line 83, in preprocess data[data_name] = self.check_and_convert(data[data_name], rows, cols, direction="to") File "/datashare3/charis/anaconda/envs/simp/lib/python3.7/site-packages/albumentations/core/utils.py", line 91, in check_and_convert return self.convert_to_albumentations(data, rows, cols) File "/datashare3/charis/anaconda/envs/simp/lib/python3.7/site-packages/albumentations/core/bbox_utils.py", line 142, in convert_to_albumentations return convert_bboxes_to_albumentations(data, self.params.format, rows, cols, check_validity=True) File "/datashare3/charis/anaconda/envs/simp/lib/python3.7/site-packages/albumentations/core/bbox_utils.py", line 408, in convert_bboxes_to_albumentations return [convert_bbox_to_albumentations(bbox, source_format, rows, cols, check_validity) for bbox in bboxes] File "/datashare3/charis/anaconda/envs/simp/lib/python3.7/site-packages/albumentations/core/bbox_utils.py", line 408, in return [convert_bbox_to_albumentations(bbox, source_format, rows, cols, check_validity) for bbox in bboxes] File "/datashare3/charis/anaconda/envs/simp/lib/python3.7/site-packages/albumentations/core/bbox_utils.py", line 352, in convert_bbox_to_albumentations check_bbox(bbox) File "/datashare3/charis/anaconda/envs/simp/lib/python3.7/site-packages/albumentations/core/bbox_utils.py", line 435, in check_bbox raise ValueError(f"Expected {name} for bbox {bbox} to be in the range [0.0, 1.0], got {value}.") ValueError: Expected y_min for bbox (tensor(0.5800), tensor(1.0111), tensor(0.7067), tensor(1.), tensor(1)) to be in the range [0.0, 1.0], got 1.0110957622528076.

It looks like, for some images, the coordinate for y_min has been annotated out of the image border. You may need to check which images have that issue.

Hello, I have just pushed an update to datasets.py that removes all files with invalid bounding boxes before training. Please check. I am closing the issue for now. Please re-open if needed.