This repo contains materials for the course ErSE328 Advanced Seismic Inversion course taught by Professor Tariq Alkhalifah at King Abdullah University of Science(KAUST).
This repository is organized as follows:
- 📂 package: Ignore
- 📂 asset: Ignore
- 📂 data: folder containing data for the assignments
- 📂 notebooks: set of Jupyter Notebook-based assignments
- 📂 scripts: Ignore
Throughout the computational part of the course, we will mainly be utilizing the Deepwave Python library which you can access from their repository. To get yourself started, you can directly learn the Deepwave fundamentals from their documentations.
To successfully install the required environment, I highly recommend using a Linux system and ensuring your machine has a GPU. (Ibex would be a great choice)
Step 1.1 Install Anaconda (if not already installed)
you can find the Linux version of Anaconda in Anaconda. Then you can install a suitable Anaconda package by running (for example I use "https://repo.anaconda.com/archive/Anaconda3-2024.10-1-Linux-x86_64.sh" )
wget https://repo.anaconda.com/archive/Anaconda3-2024.10-1-Linux-x86_64.sh
Step 1.2: Run the installer script
chmod +x Anaconda3-2024.10-1-Linux-x86_64.sh
sh Anaconda3-2024.10-1-Linux-x86_64.sh
Step 1.3: Initialize Anaconda
source ~/.bashrc
After installing the anaconda package, typing
conda -V
to testify whether you have installed Anaconda successfully
Step 2.1: To install the Deepwave environment, run the following command:
sh install_env.sh
It will take some time, but if, in the end, you see the word Done! on your terminal, you are ready to go.
Step 2.2: Remember to always activate the environment by typing:
conda activate erse328asi
First connect to Ibex using your KAUST credential.
ssh USERNAME@glogin.ibex.kaust.edu.sa
Then, clone this repository and install the erse328asi conda environment. If it is installed please go ahead with the following command. If not, please refer to this documentation to install conda. Run the following command to submit the slurm jupyter notebook request.
sbatch erse328asi_notebook.slurm
You can connect from your workstation to access the notebook with the instructions from the output slurm job. The file is in the format of slurm-JOBID.out.
ssh -L 6789:GPUID:6789 USERNAME@glogin.ibex.kaust.edu.sa
where the JOBID and GPUID are the unique identifiers from the slurm output job request. Then, access the http://localhost:6789/ link from your workstation and fill in the credential from the slurm-JOBID.out.
The following notebooks are provided:
- 📙
notebooks/01_marmousi/notebook.ipynb: First assignment (Deadline: March 6th) - 📙
notebooks/02_multiscale/notebook.ipynb: Second assignment (Deadline: March 20th)