Source code of the paper entitled as "Noise-tolerant Self-embedded LSTM for Seismic Event Classification", published on IEEE MLSP 2023.
This code was developed using Python 3.8, and you can follow the steps below for project execution:
- Create a Python virtual environment
- Install project requirements
pip install -r requirements.txt
- Create a
data/
folder, download and save the experiment dataset (originally published on https://doi.org/10.1016/j.dib.2020.105627). - Split the dataset into train and test sets to save volcano data and labels using the NPY format. We strongly recommend using the
x_y_trainset.npy
andx_y_testset.npy
names to save the train and test datasets on the data folder.
After doing all the steps, you can run python main.py
.