Autoencoders for Seismic Waveforms: Compress, Learn, and Generate
Autoencoders learn compact representations by reconstructing seismic data. This tutorial explains their use...
Utpal Kumar, PhD
Earthquakes right now
Magnitude 2.5+ over the past 24 hours.
Loading recent earthquakes…
- 149+ Posts
- 12 Topics
- Oct 2026 Updated
RLHF Explained: How Human Feedback Shapes AI Agents
How human comparisons become a training signal, where DPO fits, and why useful AI agents still need indepen...
Recent posts
Recent technical articles on seismic analytics, scientific computing, numerical modeling, and production data workflows.
How to implement the iterative newton–raphson method to find roots of a function in python
The Newton–Raphson method (commonly known as Newton’s method) is developed for finding roots of a given function or polynomial iteratively. We show two examp...
How to host a http server on your linux computer to share files
Spin up a zero-dependency web server with Python’s built-in http.server to share files or preview a site on your local network.
How effective is the signal denoising using the matlab based wavelet analysis
We see how to download seismic waveforms, convert them into mat format from mini-seed and then perform denoising using wavelet analysis. We first performed w...
Make your python script executable from anywhere in linux
Follow the instructions to make your python script executable from anywhere in Linux system.
How to securely send data from local client to remote server using paramiko (codes included)
Paramiko module can be used in Python to securely send data from the local client to the remote server. It is analogous to the SSH and SCP in Linux.
Towards multi-resolution analysis with wavelet transform
We will learn the basic concepts of wavelet tranform and multi-resolution analysis starting from the Fourier Transform, and Gabor Transform.
Empirical orthogonal function analysis to inspect the spatial coherency in the geospatial data (codes included)
Empirical Orthogonal Functions analysis decomposes the continuous space-time field into a set of orthogonal spatial patterns along with a set of associated u...
Concatenating daily seismic traces into one miniseed file (codes included)
I concatenate the daily seismic traces for 15 days into one miniseed file for further analysis. Then I obtained the spectrogram of the 15 days seismic data. ...
Computing cross-correlation and spectrogram of two seismic traces (codes included)
Read the seismic traces from the miniseed files and compute the cross-correlation and spectrogram
Principal component analysis to decompose signals and reduce dimensionality (codes included)
Use Principal Component Analysis (via SVD) to decompose a space-time signal into a few dominant modes and reduce its dimensionality, with Python code.
Signal denoising using fourier analysis in python (codes included)
We will learn the basics of Fourier analysis and implement it to remove noise from the synthetic and real signals
Analyzing miniseed seismic data in matlab (codes included)
We will learn how to convert a mseed data file into mat format and then read and analyze it using MATLAB
Transfer learning applied on the unsplash data using alexnet pretrained network (codes included)
Transfer learning using the pre-trained deep learning networks from MATLAB can be easily implemented to achieve fast and impressive results
Getting familiar with pytorch data structures (codes included)
In this introduction to the concepts of Pytorch data structures, we will learn about how to create and reshape tensors using Pytorch and compare it with the ...
How to implement the iterative newton–raphson method to find roots of a function in python
The Newton–Raphson method (commonly known as Newton’s method) is developed for finding roots of a given function or polynomial iteratively. We show two examp...
How to host a http server on your linux computer to share files
Spin up a zero-dependency web server with Python’s built-in http.server to share files or preview a site on your local network.
How effective is the signal denoising using the matlab based wavelet analysis
We see how to download seismic waveforms, convert them into mat format from mini-seed and then perform denoising using wavelet analysis. We first performed w...
Make your python script executable from anywhere in linux
Follow the instructions to make your python script executable from anywhere in Linux system.
How to securely send data from local client to remote server using paramiko (codes included)
Paramiko module can be used in Python to securely send data from the local client to the remote server. It is analogous to the SSH and SCP in Linux.
Towards multi-resolution analysis with wavelet transform
We will learn the basic concepts of wavelet tranform and multi-resolution analysis starting from the Fourier Transform, and Gabor Transform.
Empirical orthogonal function analysis to inspect the spatial coherency in the geospatial data (codes included)
Empirical Orthogonal Functions analysis decomposes the continuous space-time field into a set of orthogonal spatial patterns along with a set of associated u...
Concatenating daily seismic traces into one miniseed file (codes included)
I concatenate the daily seismic traces for 15 days into one miniseed file for further analysis. Then I obtained the spectrogram of the 15 days seismic data. ...
Computing cross-correlation and spectrogram of two seismic traces (codes included)
Read the seismic traces from the miniseed files and compute the cross-correlation and spectrogram
Principal component analysis to decompose signals and reduce dimensionality (codes included)
Use Principal Component Analysis (via SVD) to decompose a space-time signal into a few dominant modes and reduce its dimensionality, with Python code.
Signal denoising using fourier analysis in python (codes included)
We will learn the basics of Fourier analysis and implement it to remove noise from the synthetic and real signals
Analyzing miniseed seismic data in matlab (codes included)
We will learn how to convert a mseed data file into mat format and then read and analyze it using MATLAB
Transfer learning applied on the unsplash data using alexnet pretrained network (codes included)
Transfer learning using the pre-trained deep learning networks from MATLAB can be easily implemented to achieve fast and impressive results
Getting familiar with pytorch data structures (codes included)
In this introduction to the concepts of Pytorch data structures, we will learn about how to create and reshape tensors using Pytorch and compare it with the ...