Logo J. Marcus Hughes
    Logo Inverted Logo
    • Posts
    • Welcome
    • Anomaly Detection
      • Atypicality
      • Axis-aligned RRCF artifacts
      • goal of nonstationary
      • Mostly noise training
    • Assorted Coding
      • flood
      • inclusion check
      • mood tracking
      • np.sum speeds in Cython
      • sliding puzzles
      • truncating numbers
      • Ulam Warburton
    • Assorted ML
      • fake APOD generation
      • TSS vs f1
    • Denoising
      • denoising autoencoder
    • Paper Reading
      • State of Astro-informatics
    • Photography
      • New Instagram
    • Solar
      • AIA temporal coloring
      • EUI artifact
      • HEK thematic maps
      • SUVI calendar heatmap
      • Thematic Map Paper
    • Teaching
      • peer review activity
    Hero Image
    AIA temporal coloring

    Temporal coloring I downloaded a few hours of AIA data in 171 Angstroms for another project. (For the non-solar people, AIA is the Atmospheric Imaging Assembly aboard NASA’s Solar Dynamics Observatory. It takes pictures of the Sun in extreme ultraviolet.) I started playing around with it, and I ended up making the hypnotic and enchanting, at least in my opinion, video at the end of this post. It was a simple process.

    June 3, 2022 Read
    Hero Image
    Creating fake APOD pictures

    Introduction I volunteer to teach programming to two high school students through a non-profit called The Coding School, specifically the CodeConnects program. One of my students is taking AP Computer Science at the same time. His teacher requires all the students to complete a final project of their choosing; they have complete freedom in picking the project except it has to be approved by the teacher. (That sounds like an awesome teacher by the way, encouraging students to learn on their own by pursuing their own passion projects.

    May 30, 2022 Read
    Hero Image
    SUVI calendar heatmap

    Introduction The Solar Ultraviolet Imager (SUVI) aboard NOAA’s GOES takes pictures of the Sun in 6 ultraviolet passbands. They’re quite stunning and are used for space weather monitoring and prediction. (I was lucky enough to get to work with them in my previous job at NOAA.) To make them more useful, the imager takes long and short length exposures and then composites them together as a Level 2 composite image product.

    May 20, 2022 Read
    Hero Image
    EUI Artifact Analysis

    Introduction The Extreme Ultraviolet Imager (EUI) aboard the ESA/NASA Solar Orbiter mission takes pictures of the Sun in ultraviolet wavelengths. Above you can see an annotated version of one of these images. Specifically this is the image with filename solo_L1_eui-fsi174-image_20210224T013042238_V02.fits. Notice the artifacts where above and below the solar disk there are darkened stripes. It’s my understanding from a colleague that the cause of these stripes is unknown and they still need to be corrected in the data.

    April 1, 2022 Read
    Hero Image
    Efficient Inclusion Check

    Problem Statement I recently stumbled upon a cute problem where you have three sets of numbers $$A, B, C \subseteq \lbrace 0, \dots, n \rbrace.$$ You want to answer if there exists $$a \in A, b \in B \text{ such that } c = a + b \in C.$$ It’s a conceptually simple problem: are there two numbers from the first two sets that sum to a number in the third set?

    March 25, 2022 Read
    Hero Image
    Welcome

    Welcome to my new site. I hope you find interesting tidbits from my work here. Old posts Old posts can be found at my old blog. New things will be posted here from now on. I have copied many posts from the old blog.

    February 18, 2022 Read
    Hero Image
    New Instagram

    I’ve had a 500px account for some time, but I just decided to make an Instagram account. I did it mostly to enter the Denver Botanic Gardens Orchid Showcase Photo Contest, but I think it could be a fun way to share my interest in photography as well. I’ll be posting my favorite of favorite pictures at 500px, and then everything at Instagram. Hopefully, this means I’ll take more photos.

    February 16, 2022 Read
    Hero Image
    Mood tracking: first plots

    For some time now, I’ve been using the app Daylio to track my mood. It allows me to rate my mood on a 5 point scale: awful, bad, meh, good, and rad. I can also put pictures, notes, or log what activities are related to that mood at the same time. It’s quite a nifty little app. I just decided to download the data and start examining it more critically. I hope to build a small set of tools to automatically process all the data and interpret conclusions.

    February 6, 2022 Read
    Hero Image
    Sliding puzzles: my beginning

    I recently have become interested in sliding puzzles. You can read all about them on Wikipedia. In undergraduate AI, I was ask to implement a solver for the rush-hour game, a version of a sliding puzzle. I forgot all about it until recently when I was preparing an AI lesson to teach for Code Connects, a coding education non-profit. I stumbled upon the 8-puzzle. Difficulty of AI At first, it seemed pretty simple, and I thought an AI would do easy on it for bigger boards, but I was sorely mistaken.

    January 27, 2022 Read
    Hero Image
    Comparison of speed of np.sum in Cython

    I was curious how much the overhead of np.sum impacted Cython code. That is, should you write your own sum method that loops or use np.sum? So, I wrote up a little test definition of the two approaches, as shown below: import numpy as np cimport numpy as np cpdef double sum_np(np.ndarray[double] a): return np.sum(a) cpdef double sum_loop(double[:] a) nogil: cdef size_t i, I cdef double total = 0.0 I = a.

    May 11, 2021 Read
    Hero Image
    Truncate to specified number of significant figures

    It’s a simple request. You have a number, e.g. 1.25, and want to truncate it to say 2 significant figures. You specifically need to truncate instead of rounding, i.e. you need to get 1.2 instead of 1.3. How do you do that quickly in Python? Check out this code snippet: from math import log10, floor import decimal decimal.getcontext().rounding = decimal.ROUND_DOWN def truncate_sig(x, sig=2): x = float(x) # make sure it's a float or decimal throws an error if x == 0: # can't take log10(0) so just return 0 return 0 elif x <= 0: # if it's negative, determine sigfigs of positive and multiply by -1 return -1 * truncate_sig(np.

    May 10, 2021 Read
    Hero Image
    Paper published: Real-time solar image classification: Assessing spectral, pixel-based approaches

    This was my first first-author paper! It talks about doing semantic segmentation on the Sun. Here’s the link.

    November 16, 2019 Read
    • ««
    • «
    • 1
    • 2
    • »
    • »»
    Navigation
    • About
    Contact me:
    • hughes.jmb@gmail.com
    • 1-270-704-3055

    Toha Theme Logo Toha
    © 2022 Copyright James Marcus Hughes
    Powered by Hugo Logo