create_voc_dataset('12', 'train', True, True)

[ghost]

No such file or directory: '/home/cbl/PycharmProjects/blitznet-master/Datasets/VOCdevkit/VOC2012/ImageSets/SegmentationAug/train.txt'

SegmentationAug is not found in the voc12

[dvornikita]

You don't have the annotations that are getting created when you prepare the datasets. This shouldn't happen if you perform the preparation steps, as noted in the readme. Did you do that?

[ghost]

I do the step 1 2 3 ,but in the step 3,set_extra_annotations, subprocess.call('rm %s' % archive_name) meet some problem

[ghost]

but in step 3, subprocess.call('rm %s' % archive_name) meet some problem

[ghost]

I do the step 1 2 3 ,but in the step 3,set_extra_annotations, subprocess.call('rm %s' % archive_name) meet some problem

[ghost]

when I run create_voc_dataset('12', 'train', True, True) FileNotFoundError: [Errno 2] No such file or directory: '/home/cbl/PycharmProjects/blitznet-master/Datasets/VOCdevkit/VOC2012/SegmentationClass/2010_005825.png'

why?

[dvornikita]

This means that either your path to annotations is not correct, or the annotations don't exist. In the first case, you need to check the path and correct it in case it doesn't. In the latter case, you need to make sure you did all the steps correctly in order to download and unpack the dataset.

P.S. The failure of subprocess.call('rm %s' % archive_name) shouldn't influence the presence of annotations.

[Roger2148]

      You don't have the annotations that are getting created when you prepare the datasets. This shouldn't happen if you perform the preparation steps, as noted in the readme. Did you do that?

I just followed the preparation steps

#create_voc_dataset('07', 'test')                             completed
#create_voc_dataset('07', 'trainval')                       completed
create_voc_dataset('12', 'train', True, True)
#create_voc_dataset('12', 'val', True)                     completed

then I have the same problem... FileNotFoundError: [Errno 2] No such file or directory: 'C:\Users\Zhang\Downloads\blitznet\Datasets\VOCdevkit/VOC2012/ImageSets/SegmentationAug/train.txt'

was i download an imcomplete tar files? sincerely

[dvornikita]

@Roger2148 did you complete the step 3 of preparations? This is what creates the file you are missing.

[Roger2148]

@dvornikita in the step 3, The original link of the website(http://www.eecs.berkeley.edu/Research/Projects/CS/vision/grouping/semantic_contours/benchmark.tgz) has been redirected( or disappeared?), finally I saw https://github.com/vlfeat/matconvnet-fcn/issues/54 and now i just downloaded from google drive. so i think maybe the link in set_extra_annotations.py should be renewed