Mmdetection: 训练VOC数据集出错

Created on 19 Oct 2019 · 5Comments · Source: open-mmlab/mmdetection

(mmdetection) cb@ceo-pc:~/mmdetection$ python tools/train.py configs/cascade_mask_rcnn_r50_fpn_1x.py
2019-10-19 10:52:50,646 - INFO - Distributed training: False
2019-10-19 10:52:52,295 - INFO - load model from: torchvision://resnet50
2019-10-19 10:52:52,643 - WARNING - The model and loaded state dict do not match exactly

unexpected key in source state_dict: fc.weight, fc.bias

2019-10-19 10:52:58,388 - INFO - Start running, host: cb@ceo-pc, work_dir: /home/cb/mmdetection/work_dirs/cascade_mask_rcnn_r50_fpn_1x
2019-10-19 10:52:58,388 - INFO - workflow: [('train', 1)], max: 12 epochs
Traceback (most recent call last):
File "tools/train.py", line 108, in
main()
File "tools/train.py", line 104, in main
logger=logger)
File "/home/cb/mmdetection/mmdet/apis/train.py", line 60, in train_detector
_non_dist_train(model, dataset, cfg, validate=validate)
File "/home/cb/mmdetection/mmdet/apis/train.py", line 227, in _non_dist_train
runner.run(data_loaders, cfg.workflow, cfg.total_epochs)
File "/home/cb/anaconda3/envs/mmdetection/lib/python3.6/site-packages/mmcv/runner/runner.py", line 358, in run
epoch_runner(data_loaders[i], **kwargs)
File "/home/cb/anaconda3/envs/mmdetection/lib/python3.6/site-packages/mmcv/runner/runner.py", line 260, in train
for i, data_batch in enumerate(data_loader):
File "/home/cb/anaconda3/envs/mmdetection/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 582, in __next__
return self._process_next_batch(batch)
File "/home/cb/anaconda3/envs/mmdetection/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 606, in _process_next_batch
raise Exception("KeyError:" + batch.exc_msg)
Exception: KeyError:Traceback (most recent call last):
File "/home/cb/anaconda3/envs/mmdetection/lib/python3.6/site-packages/torch/utils/data/_utils/worker.py", line 99, in _worker_loop
samples = collate_fn([dataset[i] for i in batch_indices])
File "/home/cb/anaconda3/envs/mmdetection/lib/python3.6/site-packages/torch/utils/data/_utils/worker.py", line 99, in
samples = collate_fn([dataset[i] for i in batch_indices])
File "/home/cb/mmdetection/mmdet/datasets/custom.py", line 128, in __getitem__
data = self.prepare_train_img(idx)
File "/home/cb/mmdetection/mmdet/datasets/custom.py", line 141, in prepare_train_img
return self.pipeline(results)
File "/home/cb/mmdetection/mmdet/datasets/pipelines/compose.py", line 24, in __call__
data = t(data)
File "/home/cb/mmdetection/mmdet/datasets/pipelines/loading.py", line 107, in __call__
results = self._load_masks(results)
File "/home/cb/mmdetection/mmdet/datasets/pipelines/loading.py", line 86, in _load_masks
gt_masks = results['ann_info']['masks']
KeyError: 'masks'