@jdszekeres

I fixed the video demo to show in a reasonable amount of time and built a demo cli with the sunglasses filter as an exe.

I built a program that makes a cool animation to show before and after the filter

A good model run testing with 28 photos with my sunglasses as the filter: it’s finally scoring well ~88% and actually looks like it should.

I’ve also worked on a cli tool to work with the ml_model output to apply filters, view histograms, list settings, and score models

first full model training and it went TERRIBLE. My scoring was based on the histogram for the image, so color is not preserved . These photos have very similar average light values. I’m going to rerun the model with a new scoring algorithm that scores each color channel individually.

I added a pixelation load in effect to my video, so it looks like its getting clearer and clearer as it loads in.

I worked on the landing page of my personal website and I added a bunch of photos of myself that play

I’ve added the first draft of alarm clock code. It fetches time over NTP and plays the Monty Python Lumberjack song at alarm time (Because it plays right after the Dead Parrot Sketch in “And Now for Something Completely Different”).

Completed what is hopefully my final design for the enclosure. It features both the buttons and screen on the top, which differs from the guide, but don’t worry, It’s all going to be turned sideways in the end. Also, if you look very closely, you’ll see that my buzzer hole is in the shape of a parrot that is resting/pining for the fjords

Hi, Hola, Bonjour, and welcome to my Blare devlogs. During this first session I’ve created my PCB and KiCad and imported it into Onshape.

To make my project unique I am going to add a web control interface as well and I’ve added clipart and quotes from Monty Python Dead Parrot Scene to the PCB silkscreen

Add celestial objects to FOV preview

Changelog

• 2a13b7a

I fixed the Bortle Scale by building my own light pollution map from scratch using a light pollution scattering kernel using the CU Boulder VIIRS dataset.

Previously, the dataset I used didn’t have enough labels for each bortle scale and had country borders, which screwed with what the expected light pollution is.

I’ve taken your feedback from the last ship and I’ve started work on a more space-y theme for this project. So far, I’ve switched up the color scheme to a dark blueish-purple and comsic latte (The average color of the universe). I also added a procderual starfield with comets in the background. Also, check out the homepage, the page cards also float around like they’re in microgravity!

I did a lot of space saving (pun intended) to reduce the offline cache size by about 0.7 Mb to 3.8 Mb by better compressing images and minifying code and JSON. I also had to do an emergency fix for mobile, where it wasn’t working on certain phones.

Added a guide to Astrophotography tools to make it more beginner friendly.

Added milky way to star chart! It looks kinda cool how it wraps off the page and it definitely wasn’t because it was too hard to remove it.

I added a FOV preview, which shows how the 110 messier objects appear in photos based on your camera sensor size and lens focal length. It’s very useful for deep space imaging.

Test model done on basic settings (Averaging around 92% accuracy, which is good, but it did it in a completely different way)

Next I’m going to train it on a more complex filter