search

chanyn / HKRM

Hybrid Knowledge Routed Module for Large-scale Object Detection (NerIPS2018)
105 stars 19 forks source link

it occurs a error after I train the model about thousands of iterations,do you know why #10

Closed summerrr closed 5 years ago

summerrr commented 5 years ago

[session 2][epoch 1][iter 3800] loss: 1.4599, lr: 5.00e-03 fg/bg=(64/192), time cost: 116.947137 rpn_cls: 0.2889, rpn_box: 0.2576, rcnn_cls: 0.8537, rcnn_box 0.3310, adja_loss 0.1700, adjr_loss 0.0107

it seems right,but it happens a error, I don't know the reason.

Traceback (most recent call last): File "trainval_HKRM.py", line 364, in data = next(data_iter) File "/home/user10/anaconda3/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 281, in next return self._process_next_batch(batch) File "/home/user10/anaconda3/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 301, in _process_next_batch raise batch.exc_type(batch.exc_msg) IndexError: Traceback (most recent call last): File "/home/user10/anaconda3/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 55, in _worker_loop samples = collate_fn([dataset[i] for i in batch_indices]) File "/home/user10/anaconda3/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 55, in samples = collate_fn([dataset[i] for i in batch_indices]) File "/home/user10/notebook/HKRM/lib/roi_data_layer/roibatchLoader.py", line 72, in getitem blobs = get_minibatch(minibatch_db, self._num_classes, self.random_scale_inds) File "/home/user10/notebook/HKRM/lib/roi_data_layer/minibatch.py", line 30, in get_minibatch im_blob, im_scales = _get_image_blob(roidb, random_scale_inds) File "/home/user10/notebook/HKRM/lib/roi_data_layer/minibatch.py", line 75, in _get_image_blob im = im[:,:,::-1] IndexError: too many indices for array

chanyn commented 5 years ago

Maybe some images are missing, so "im" is valid. Please check your dataset.