dijkprofile-annotator 0.1.1

Automatically annotate dijkprofiles in qDAMEdit format

Version: 0.1.1

dijkprofile-annotator description

Automatically annotate drijkprofile is qDAMEdit format

Author: Jonathan Gerbscheid
Email: [e-mail]

Online Tool

An web version of this tool is available at: https://huggingface.co/spaces/jgerbscheid/dpa-example
The availability of this tool is not currently guaranteed and it's location might change in the future to a different adress.

Install

Warning

This package will install pytorch to run the neural network for prediction. If you wish to use your own pytorch installation or modify the code in any way I recommend cloning the repository and installing locally:
https://gitlab.com/hetwaterschapshuis/kenniscentrum/tooling/dijkprofile-annotator/-/tree/master/
I recommended installing the package in a fresh conda environment to avoid conflicts with already other installed packages.

Install directly from PiPI with pip

pip install dijkprofile-annotator

Installing locally:

git clone [e-mail]:hetwaterschapshuis/kenniscentrum/tooling/dijkprofile-annotator.git
cd dijkprofile-annotator
pip install -e . 

Usage

basic

To annotate a single file, use the annotate function:

import dijkprofile_annotator

input_filepath = "/home/documents/surfacelines.csv"
target_filepath = "/home/documents/predicted_characteristpoints.csv"

dijkprofile_annotator.annotate(input_filepath, 
                               target_filepath)

Detailed Exampes

See the example notebooks at:
https://gitlab.com/hetwaterschapshuis/kenniscentrum/tooling/dijkprofile-annotator/-/tree/master/notebooks
for examples on how to use the package.