Issue running parameter files inWSL2

[KritiJainPhd]

I am new to using LSDTopoTool and trying to run it in WSL2 Ubuntu 22.04LTS terminal window using Docker. When I try to run the lsdtt-basic-metric or other driver files, the output says there is no such files in the directory but I have the example datasets in the same directory where LSDTopoTool is installed. Maybe the Docker container is not properly linked to the directory in the host system. Please suggest me how to fix this.
Thank You
Kriti[

[simon-m-mudd]

Hello Kriti
If you are running in Ubuntu within WSL you need to use the Linux binding:
docker run -it -v /LSDTopoTools:/LSDTopoTools lsdtopotools/lsdtt_opencv_docker
It is better to run the docker container directly using docker for windows rather than within a wsl ubuntu distribution, since you will need to mount your file system twice (once to get your ubuntu system to see your windows files, and again to get the docker container to see the ubuntu files).

That docker container is somewhat out of date.
For the time being I suggest running the code on google colab.
Many examples are here: https://github.com/LSDtopotools/lsdtt_notebooks/tree/master/lsdtopotools

I am in the process of migrating the containers to apptainer. There will be new documentation for this online soon.

You can install apptainer in the ubuntu distribution with

$ sudo add-apt-repository -y ppa:apptainer/ppa
$ sudo apt update
$ sudo apt install -y apptainer
$ sudo add-apt-repository -y ppa:apptainer/ppa
$ sudo apt update
$ sudo apt install -y apptainer

Then you can download the container here:
$ wget https://www.geos.ed.ac.uk/~smudd/apptainer_containers/lsdtt_distrib.sif

and run with
$ apptainer bash lsdtt_distrib.sif

[KritiJainPhd]

Thank you @simon-m-mudd for the quickest response. I am currently able to run docker using docker for windows to run Morphometric Analysis. Meanwhile I wait for availability for documentation of the migrated containers to apptainer, according to you what will give me the better results using docker for windows or running the code within google colab.
Kriti

[simon-m-mudd]

Hello @KritiJainPhd , google colab has very fast CPUs but the main issue is file transfer. If you download DEMs from opentopography (instructions are in the notebooks) it will be fast. But if you have your own DEMs on your local computer that you want to transfer to colab it will be slow.
So if you just want to do exploration using global DEMs I would use colab. If you want to retain all data and records of analysis for a paper use docker/apptainer on your local computer.

[KritiJainPhd]

Thank You so much for your suggestion.
Kriti