search

jwyang / faster-rcnn.pytorch

A faster pytorch implementation of faster r-cnn
MIT License
7.67k stars 2.33k forks source link

AssertionError: Path does not exist: /media/user1/storage-1/Devendra/faster_rcnn_2_jwyang/faster-rcnn.pytorch/data/VOCdevkit2007/VOC2007/JPEGImages/filename #717

Closed devendraswamy closed 4 years ago

devendraswamy commented 4 years ago

/media/user1/storage-1/V4_DATASET/new_annotation/train_person.txt data loaded /media/user1/storage-1/V4_DATASET/new_annotation/test_person.txt data loaded

/media/user1/storage-1/Devendra/faster_rcnn_2_jwyang/faster-rcnn.pytorch/data/VOCdevkit2007/VOC2007 data path Loaded dataset voc_2007_train for training Set proposal method: gt Appending horizontally-flipped training examples... Traceback (most recent call last): File "trainval_net.py", line 203, in imdb, roidb, ratio_list, ratio_index = combined_roidb(args.imdb_name) File "/media/user1/storage-1/Devendra/faster_rcnn_2_jwyang/faster-rcnn.pytorch/lib/roi_data_layer/roidb.py", line 116, in combined_roidb roidbs = [get_roidb(s) for s in imdb_names.split('+')] File "/media/user1/storage-1/Devendra/faster_rcnn_2_jwyang/faster-rcnn.pytorch/lib/roi_data_layer/roidb.py", line 116, in roidbs = [get_roidb(s) for s in imdb_names.split('+')] File "/media/user1/storage-1/Devendra/faster_rcnn_2_jwyang/faster-rcnn.pytorch/lib/roi_data_layer/roidb.py", line 113, in get_roidb roidb = get_training_roidb(imdb) File "/media/user1/storage-1/Devendra/faster_rcnn_2_jwyang/faster-rcnn.pytorch/lib/roi_data_layer/roidb.py", line 97, in get_training_roidb imdb.append_flipped_images() File "/media/user1/storage-1/Devendra/faster_rcnn_2_jwyang/faster-rcnn.pytorch/lib/datasets/imdb.py", line 117, in append_flipped_images widths = self._get_widths() File "/media/user1/storage-1/Devendra/faster_rcnn_2_jwyang/faster-rcnn.pytorch/lib/datasets/imdb.py", line 113, in _get_widths for i in range(self.num_images)] File "/media/user1/storage-1/Devendra/faster_rcnn_2_jwyang/faster-rcnn.pytorch/lib/datasets/imdb.py", line 113, in for i in range(self.num_images)] File "/media/user1/storage-1/Devendra/faster_rcnn_2_jwyang/faster-rcnn.pytorch/lib/datasets/own_dataset.py", line 86, in image_path_at return self.image_path_from_index(self._image_index[i]) File "/media/user1/storage-1/Devendra/faster_rcnn_2_jwyang/faster-rcnn.pytorch/lib/datasets/own_dataset.py", line 102, in image_path_from_index 'Path does not exist: {}'.format(image_path) AssertionError: Path does not exist: /media/user1/storage-1/Devendra/faster_rcnn_2_jwyang/faster-rcnn.pytorch/data/VOCdevkit2007/VOC2007/JPEGImages/filename

please help me to solve this error, I am try to training with my own dataset .

ElsaReedz commented 4 years ago

I have the same problem, did you solve it?

devendraswamy commented 4 years ago

That is the path error , i solved that error by adding my own file path , if your using pascal_voc dataset then please check dataset path and if it is not there then arrange based upon the error.

ElsaReedz commented 4 years ago

That is the path error , i solved that error by adding my own file path , if your using pascal_voc dataset then please check dataset path and if it is not there then arrange based upon the error.

I solved that too, but cant remember the detail, like found myself creating soft link in the wrong path I suppose, seemingly the same way

devendraswamy commented 4 years ago

I thought soft link was not require and on which data you kept the training ? custom or pascal_voc

ElsaReedz commented 4 years ago

I thought soft link was not require and on which data you kept the training ? custom or pascal_voc

pascal_voc, and I do remember when I encountered with this, I followed the tuturial in https://github.com/rbgirshick/py-faster-rcnn#beyond-the-demo-installation-for-training-and-testing-models and tried create the soft link VOCdevkit2007 under /VOCdevkit, but failed to include the whole path so I created that in the wrong place, anyway I completed the path and this error no longer comes out again

devendraswamy commented 4 years ago

I need a one help , when i am kept the training with large data then my system was restarted. why its happening and its works for less data.

please help me to find the solution

ElsaReedz commented 4 years ago

I need a one help , when i am kept the training with large data then my system was restarted. why its happening and its works for less data.

please help me to find the solution

I ran the whole project on a cloud server, trained with a large dataset of over 10k images while also a much smaller one containing only 270 which is made by my team, and did not encounter with such issue, so I may not able to be of help. I found some of my fellows met the same problem however, one of them checked his hardware and solve it issue: https://blog.csdn.net/wuzuyu365/article/details/52212544 summary of related: https://blog.csdn.net/u011622208/article/details/90579252 hopefully it would help

devendraswamy commented 4 years ago

Thank you , the blogs are suggested to increase the power supply to GPU. ok i will try and if any other solution please drop here.

devendraswamy commented 4 years ago

File "/media/user1/storage-1/Devendra/TF_model/tf-faster-rcnn/tools/../lib/datasets/voc_eval.py", line 251, in voc_eval if not R['det'][jmax]: IndexError: list index out of range

please help ,Thanking you!!!