python 2.7 - how to create & encode a label file using sample images for tensorflow -

we trying detect special characters + , - inside image using tensorflow extending mnist sample code -> https://github.com/opensourcesblog/tensorflow-mnist

we have been able create binary encoded file using our sample images needed training neural network using sample code -> https://github.com/jkarnows/idx-formatter

but not finding way how create label file our images , create binary encoded label file using label file

both of these files important proceeding further .

anybody having idea welcome share them

the data in gzip format, decoded extract_labels, each label on 8 bytes. can use numpy.getbuffer convert uint8 array back.

alternatively, can create own extract_labels reads labels in whatever format see fit.