git clone https://gitlab2.informatik.uni-wuerzburg.de/chr58bk/mptv.git
python -m pip install --user virtualenv
python -m virtualenv path/to/venv
source path/to/venv/bin/activate
pip install -r requirements.txt
python setup.py install
Input: image list (*.png, *.bin.png, *.nrm.png, + corresponding GT (*.gt.txt))
Output: folder (--output, default: ./Data) with n (--folds, default = 5) sub folders which contain #lines/n lines
The folder id gets placed in front of the line name in order to prevent overwriting of lines with the same name.
Input: folder containing the "Folds" sub folder.
Output: "Models" folder which containes the trained models for each fold.
--models: "LH,LH,FRK,ENG,ENG" means that the first two folds starts training from LH fold 3 from ENG, and fold 4 and 5 from FRK with LH, ENG,
and FRK being different mixed models from the pretraining folder.
"LH,,,,LH" starts the training of fold 1 and 5 from LH and trains the rest from scratch.
Input: folder containing "Folds" and "Models" sub folders.
Output: the best model of each fold stored in "Models/BestModels".
Input: folder(s) containing the to-be-predicted lines.
Output: folder "Voting" in each input folder which contains the results (*.txt and *.extLlocs) for each fold.
Input: "Voting" folder(s) resulting from the preceding step.
Output: "Voting/Voted" folders containing voted *.txts.