blog

Pigment and structural color

Today when I listened to the podcast about “blueberry is not blue”, I recalled the question I heard a few years ago “what is the difference between colloid color and dye color”. Even to this day, I still don’t have a solid understanding of the physics of color, but rather some scattered pieces of knowledge with terminologies that do not necessarily make sense, such as Rayleigh scattering and absorption by chemical bonds. But the question is still there: is there a unified framework to understand color?
Principle Componenets Analysis

I have tried to learn machine learning a few times. However, I never find the chance to implement any machine learning algorithm in my research. Here, I want to revisit the subject by reading Goodfellow’s book. I like simple things, because they help me understand the basic concept and gain intuition to the abstract math. Principle componenets analysis is such a simple thing, which I shall take some notes here to try to understand.
How to construct pandas DataFrame efficiently

Today, I was reviewing my droplet tracking code and saw an interesting way to construct pandas.DataFrame:
My Competencies

While searching for a job, I find it useful to outline my competencies and weaknesses, so that I can quickly compare with the qualifications of the jobs and assess my fit. The content of this post would be very much aligned with a cover letter of a job application. Hence, it could be reused in the near future!
Jekyll

Recently, I start to reorganize my personal website with Jekyll. I knew this tool from the first day I tried GitHub Pages (perhaps in 2018). At that time, I did not read through the documentation and thought this is only a “theme” tool, which modifies my website to a look I don’t like. So I did not use it, but started from scratch to build an HTML + CSS static site. It was a happy learning process. I started to learn using the freecodecamp online app, and got good understanding of the web development basics. With some practice, I eventually was able to tune my website front page to the way I wanted.
Mold On Wall

Here are some pictures of the apartment at 24 Boulevard Arago Paris 75013 during our stay, mainly taken in November and December 2021. The pictures show that there was a large fraction of wall surface covered by green and black mold colonies, likely caused by high humidity. We open all the windows for about 8 hours in the day, despite the cold weather in November, and use extra electricity to heat up the chamber, the mold problem persisted. We suspect that the apartment we stayed was not properly moisture proofed. Consequently, the walls of the apartment (at least the wall paper) were damaged and the apartment was humid, cold and had unpleasant smells, hence not suitable for accommodate people.
Mendeley To Zotero

I quit Mendeley today, after 5 years of use, because there are serious problems with duplicated files, lack of support for mobile platforms, small cloud drive and poor syncing quality. Solving these problems has been distracting me from focusing on reading, so that I decide to try out some alternatives and leave Mendeley.
Personal Server

Since 2013, when I was an undergrad working on nanoparticle simulations, I was fascinated by sending the code to a server cluster to execute. The cluster was not faraway at all, like most servers in the world we are interacting with. Instead, they were just ~20 normal computers sitting in a small chamber in our office. They were running all day and all night and produce results silently.
Focusing Prioritizing And Being Realistic

How many things can I do a day? I used to think I can read two papers in the morning. In the mean time, I can prepare the experiment in the afternoon. After the experiment, I can analyze all the data with my “miraculous” code with just a click. I can even have fun in the evening! It looks like, I can have such a “fruitful” day as a researcher, without spending much time. However, this was just a daydream which I had from time to time in the past several years.
Linux Study Notes

Linux has two main branches: Debian and RedHat. Debian is meant for personal use, with a popular example Ubuntu. RedHat is meant for server use, with a popular example of CentOS. I’m going to learn the basics of servers, so the distro of interest here will be CentOS. The reference of this note is this video course. This course is intent for server setup.
Energy Spectrum 2 Methods

We have two methods for computing energy spectrum from spatial velocity data. In this post we should how they are related and point out quantitative details when converting between the results from these methods.
Gnf Evolution

Giant number fluctuation (GNF) is a key distinction of active matter compared to systems in equilibrium. Theories have predicted the deviation of number fluctuations in active matter in a quantitative way. Later on, the predictions are tested in experiment and were mostly true. Despite the success of the theories, the underlying mechanism of GNF still remains to be a hypothesis due to the difficulty of resolving individual active units and the interaction between individual active units in a crowded collection. A recent work by Peng et al. shows that in the transition to active turbulence, flow order and flow energy grow with different time scale. GNF also grows during the course of active turbulence transition, by look at the time scale of this transition, we could gain more insight into the mechanism of GNF.
Chiral Overview

Chiral particles drift in shear flows because of their asymmetric shape. A simple case that helps to understand this drift is provided by Marcos et al. (PRL 2009) where a helical particle is aligned with a simple shear flow.