The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Back to feedPost
Logistic Regression from Scratch for Pneumonia Detection in Chest X-Rays
Built a binary classification model from scratch (no ML libraries —
implemented gradient descent and the loss function manually) to detect
pneumonia from chest X-ray images, for a Numerical Methods course project.
Preprocessing: converted each X-ray into a 32x32 grayscale image,
flattened into a normalized vector, and balanced the training set to
have equal numbers of NORMAL and PNEUMONIA cases to avoid class bias.
Model: binary classifier using a tanh-based activation function,
trained via gradient descent with manually derived partial derivatives
(no autograd).
Process: tested 5 different learning rates (alpha = 0.0001 to 0.5) to
find the best trade-off between convergence speed and stability,
tracking MSE on both train and test sets across iterations.
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Trending
retrofuturism
Share your favorite designs from the past exploring the distant future. Or your own take on the aesthetic 🚀
halftone
Dots. Texture. Grain. Let's see your designs using the halftone technique.
Midjourney
Creatives are showing off what they're making with Midjourney.
colorgrade
Video editors sound off: How do you enhance the mood of your video in color grading?
Related posts
Hi everyone!
I'm new here, so I wanted to introduce myself.
I learned the basics of Python about two years ago, and a year ago I started a degree in AI and Data Science. I'm currently in my third semester, and I'm focused on developing my programming and data skills while building my first real-world experience.
For now, I'm looking to contribute to smaller tasks such as data cleaning, data analysis, automation, and logical scripting. I don't expect to jump directly into complex projects; my goal is to help with the work I can handle while learning how software and data projects actually work in the industry.
I'm also very interested in how AI fits into programming. In my personal projects, I try to use AI primarily as a learning and reasoning tool rather than as a replacement for understanding or writing the code myself. I like using it more as a Socratic teacher: asking questions, challenging my reasoning, explaining concepts, and helping me find mistakes.
At the same time, I recognize that AI is becoming almost indispensable in modern development. But I believe that understanding the fundamentals is just as important. I want to learn how things work, not simply how to get an AI to produce them.
Currently, I'm working on a physics simulator as a personal project. It's still a work in progress, but I hope to share my progress here in the future.
I'm here to learn, contribute, and get a better understanding of what working in the industry is actually like. If there are small tasks where I can be useful while continuing to learn, I'd be happy to help.
Best regards!
Built an AI-powered Log Analyzer using Python and FastAPI to automate log analysis and troubleshooting. The system processes application/system logs, performs rule-based analysis, and uses an LLM to identify problems, root causes, severity, possible causes, and recommended actions. Exposed the functionality through REST APIs with structured Pydantic responses for reliable and consistent results.
Tech: Python, FastAPI, LLM/Ollama, Pydantic, REST APIs, Log Parsing, Prompt Engineering, Git.
Here is a professional English translation of your post that captures the exact same message and tone:
Just wrapped up the project, and I realized this is definitely the picture worth sharing the most! 😁
Sometimes people look at the final dashboard and think that’s where all the work happens, but the real magic starts right here... the Data Model.
Behind every single relationship here is a lot of thinking, testing, and tweaking until everything finally clicked into place.
Alhamdulillah, the project is officially complete, and I’ve learned way more from it than I initially expected.
Any feedback is more than welcome! 🙏
#PowerBI #DataAnalytics #PowerBIDeveloper #Portfolio #OpenToWork
