Reproducible material for Deep learning based extraction of surface wave dispersion curves from seismic shot gathers - Chamorro D., Zhao J., Birnie C., Staring M., Fliedner M., Ravasi M. submitted to Near Surface Geophysics.

Project structure

This repository is organized as follows:

• 📁 data: folder containing data used for training/testing
  • 📂 testing: folder containing synthetic data generated for testing
  • 📂 training: folder containing synthetic data generated for training
  • 📂 test_example: folder containing synthetic data generated for a 2D model used to test the peformance of the inversion
• 📂 notebooks: set of jupyter notebooks reproducing the experiments in the paper as well as the two models trained (see below for more details);

Notebooks

The following notebooks are provided:

• 📙 1_dataset_generation.ipynb: notebook used to generate the training dataset
• 📙 2_training.ipynb: notebook performing the training of the network
• 📙 3_inferencing.ipynb: notebook performing the inference on real and synthetic dataset
• 📙 4_uq_analysis.ipynb: notebook performing the UQ using dropout and statistical parameters estimation methods
• 📙 5_inversion.ipynb: notebook to perform inversion using evolutionary algorithms

Disclaimer: All experiments have been carried on a Intel(R) Xeon(R) CPU @ 2.10GHz equipped with a single NVIDIA GEForce RTX 3090 GPU. Different environment configurations may be required for different combinations of workstation and GPU.