Latest Posts
-
A quick update
It’s been a while since I’ve posted anything (it was a busy summer!) so I wanted to make a quick update about some things I’ve been working on. Watch for upcoming posts on the following topics:
read more ...
-
Analytics: Asking the right questions
-
New ThinkPad!
I recently purchased my first new computer in 9 years, and I got a Lenovo ThinkPad P1 Gen 3! My previous laptop was a 2012 MacBook Pro I bought halfway through graduate school. The MBP was a great laptop and it still functions, but it has defintely started to show its age. I swapped out the internal hard drive a few years back with a solid-state drive, but with only 4 GB of memory and a tired battery, it was definitely becoming difficult to use as an effective tool.
read more ...
-
PySeisTuned
PySeisTuned2.0 is a Flask web app that allows users to produce seismic tuning wedge forward models (more on that
). Originally, I built PySeisTuned as a GUI application using Python 3 and PyQT5 because I wanted to deploy code from a Jupyter Notebook into something re-usable with minimal end-user configuration. Once I completed that initial version, I realized that deploying the software as a web app would make it even more accessible. This led me to learn Flask as it seemed like a lightweight framework perfectly suited for a simple web app. Instead of using PyQT5 widgets to build the GUI, the web app version uses HTML & Bootstrap CSS along with some custom CSS and a little JavaScript. I previously had experience building a static web page with HTML and CSS when I built a web site for my advisor’s research group AASPI while in graduate school (10 years ago now!). Building PySeisTuned2.0 was a nice refresher with HTML and CSS. read more ...
-
Building the SonicPredict API
SonicPredict is a Flask API that serves a fitted ML-model from which users can predict sonic well logs. The project began in March 2020 when I participated in the Society of Petrophysicists & Well Log Analysts (SPWLA) Petrophysical Data-Driven Analysis (PDDA) Special Interest Group’s first machine learning contest. The objective of the contest was to build a machine learning model that given a set of input well logs could predict P- and S-Sonic well logs. The contest scoring was based upon minimizing root mean squared error (RMSE) between predicted and real P- and S-Sonic well logs from a blind well not previously seen by the model. I will write more about my experience participating in the contest and the model I created in a separate post, but for now I want to write about the actual SonicPredict API.
read more ...
-
Welcome to bendowdell.com!
Welcome to my website! You can learn more about me, what I do, and how I do it on the about page. Check out some of my projects on my portfolio and take a look at my resume for more on my skills and background.
read more ...