Version: 0.1.2

dijkprofile-annotator description

Automatically annotate drijkprofile is qDAMEdit format

Author: Jonathan Gerbscheid
Email: j.gerbscheid@hetwaterschapshuis.nl

Online Tool

A 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 git@gitlab.com:hetwaterschapshuis/kenniscentrum/tooling/dijkprofile-annotator.git
cd dijkprofile-annotator
pip install -e . 

Usage

basic

After installation you can use the command line interface to annotate a single file, call the annotator with the to be labeled file and desired output file:

dijkprofile_annotator -i inputfile -o target_outputfile

It also possible to use the web interface, the following command will start a gradio app that can be accessed in a browser:

dijkprofile_annotator-gui

You can also import the module in a python script and call it from there:

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.