search

MarvinTeichmann / KittiSeg

A Kitti Road Segmentation model implemented in tensorflow.
MIT License
910 stars 403 forks source link

_load_gt_file too many values to unpack #37

Closed StuvX closed 7 years ago

StuvX commented 7 years ago

I feel like a chump, but I have produced groundtruth images along with their natural counterparts and produced a train.txt and val.txt that appear to be the same format as the kittiseg versions. When I run train.py I get the following error:

File "train.py", line 131, in tf.app.run() File "/Users/wnas0001/anaconda3/envs/python2/lib/python2.7/site-packages/tensorflow/python/platform/app.py", line 44, in run _sys.exit(main(_sys.argv[:1] + flags_passthrough)) File "train.py", line 127, in main train.do_training(hypes) File "incl/tensorvision/train.py", line 393, in do_training modules['input'].start_enqueuing_threads(hypes, queue, 'train', sess) File "/../inputs/kitti_seg_input.py", line 355, in start_enqueuing_threads gen.next() File "/../inputs/kitti_seg_input.py", line 155, in _make_data_gen for image, gt_image in data: File "/../inputs/kitti_seg_input.py", line 114, in _load_gt_file image_file, gt_image_file = file.split(" ") ValueError: too many values to unpack

I believe it has something to do with my txt file, which I've attached. Any help is greatly appreciated! train.txt

MarvinTeichmann commented 7 years ago

Yes, it seams to have to do with your data file. Maybe an extra new_line at the end of file? The error would happen, if there is a line with more then two white spaces. Try

print(" File: {}".format(file))

To see which throws the error line fails.

StuvX commented 7 years ago

Ah, yes there is something funny about the txt file I wrote in atom - I've simply edited your train3.txt and val3.txt files and it seems to work now.

StuvX commented 7 years ago

...although it ends and reports "Killed 9"...?