search

orobix / retina-unet

Retina blood vessel segmentation with a convolutional neural network
1.26k stars 468 forks source link

errors in Windows+tensorflow_GPU environment #67

Open IronLavender opened 5 years ago

IronLavender commented 5 years ago

Hello, Your experiment is Ubuntu？my expereimen is Windows+tensorflow_GPU, but it occurs error when I run the retinaNN_training.py： Traceback (most recent call last): File "D:/Study/CODES/Medical_Segmentation/Keras_retina-unet/src/retinaNN_training.py", line 145, in <module> path_data = config.get('data paths', 'path_local') # 数据文件封装后的文件路径 File "E:\Learning\Anaconda3\envs\tensorflow_gpu\lib\configparser.py", line 781, in get d = self._unify_values(section, vars) File "E:\Learning\Anaconda3\envs\tensorflow_gpu\lib\configparser.py", line 1141, in _unify_values raise NoSectionError(section) configparser.NoSectionError: No section: 'data paths' but in configurations.txt, I think it's right : [data paths] path_local = ./DRIVE_datasets_training_testing/ train_imgs_original = DRIVE_dataset_imgs_train.hdf5 train_groundTruth = DRIVE_dataset_groundTruth_train.hdf5 train_border_masks = DRIVE_dataset_borderMasks_train.hdf5 test_imgs_original = DRIVE_dataset_imgs_test.hdf5 test_groundTruth = DRIVE_dataset_groundTruth_test.hdf5 test_border_masks = DRIVE_dataset_borderMasks_test.hdf5 I fail to correct it, so this does not work in a window environment? Thank you for your work!

NeoKun004 commented 5 years ago

it does work in windows environment but there is a config file needs to be created and located to force the training to run on theano backenend, here is my email I would love to help u through this my email: ahmed.beji004@gmail.com

boba1234 commented 4 years ago

Hi, i am having this NosectionError: No section: 'data paths' too. Is there any solution to it? Thanks!

Hridoy-bit commented 4 years ago

Hi, I have got the same error . configparser.NoSectionError: No section: 'data paths'. Any solution?I emailed you NeoKun004.