Physics-Informed Neural Networks
The purpose of this post is to summarize different techniques used in the design of physics-informed neural networks (PINNs). Although advanced topics in Mac...
Leonardo Ayala
I’m a physicist turned ML engineer, with detours through chemistry, teaching, and more along the way. I write for the love of it. Mostly about science, but also about many random things that I find interesting.
PhD life: things I had to learn and unlearn
Image from storyset at Freepik This page was first posted on January 23rd of 2023, and later modified on August 20th of the same year. Mixing vibrant elixi...
I contain multitudes
Six more months to go, and I will finally get away from this school. I should know what studies to pursue, shouldn’t I? At least everyone else seems to think...
The way we sleep
If I go to sleep now, I will get six hours of sleep … Expressions like this are all to common, and lead to one of the most self-destructive behaviours that s...
Monte Carlo approximations
Monte Carlo methods is a general term of algorithms that leverage sampling strategies to approximate solution to problems that might be too computationally ...
Favourite quotes
Quotes are usually something that speaks to everyone in one way or another. I have collected the quotes that I have read over the years and that speak to me...
Favourite books
This is a collection of books that I really enjoyed and that at some point I wished I had known them sooner. I hope that you enjoy them as much as I did. Ps...
Dynamic plotting in Python (Plotly, Bokeh, mpld3 & HoloViews)
Data visualization is becoming one of the most important tools when analyzing large datasets. A simplistic rendering of a PNG image is no longer enough. Imag...
Plotly continuous error band lineplot
For some time I have been using plotly_express.line in order to generate dynamic plots. I personally think that this is a great tool that has an API very sim...