This code creates dataset to feed into caffe for classification.
How to run it
step 1: creating the train/test/validation dataset. Follow below structure: datasets- -train -test -validation
use dataset_creator.py to generate these folders. dataset_creator.py also creates train.txt, test.txt, validation.txt, these file will be used in the next step to create lmdb dataset (used in caffe models). train.txt, test.txt and validation.txt moved to caffe/data/scane-regression folder.
step 2: create lmdb dataset for caffe models - create/modify create_scaneregression.sh [need to be uploaded] - check comments to modify path such as "datasets" location - execute create_scaneregression.sh - ./create_scaneregression.sh
this step will create lmdb files for model training. lmdb files will be stored in caffe/example/scane-regression folder.