← Feed

Posts

• Post #3261479

  Excited to open-source perfusio v0.1.0! A Python toolkit for Bayesian experimental design & digital twins in perfusion bioprocessing — built for CHO cell culture, GP surrogates, and active learning loops. 📦 pip install perfusio 🔗 http://github.com/lynchaos/perfusio 🔗 https://pypi.org/project/perfusio/ 🔗 https://lynchaos.github.io/perfusio/ #bioprocess #MLOps #openscience #Python

• Post #3261478

  I have to renew all my computers. A new Linux pc, a Mac and a desktop. Too expensive 💻

• Post #3261477

  For yesterdays rally at London, I built a crowd counting system from scratch on Apple Silicon 🍎 Trained a custom Swin-T + CNN hybrid on ShanghaiTech Part A → MAE 237.69. Then tested pre-trained DM-Count models on an aerial rally photo to see how domain matters: 📸 rally.jpeg results: • DM-Count QNRF (high-res diverse): 2,077 I will soon publish this on GitHub. All running on MPS (no CUDA) 🔥 #ComputerVision #DeepLearning #CrowdCounting #PyTorch #MachineLearning

• Post #3261476

  I cloned a block puzzle for my daughter and fell down a polish rabbit hole. 🍪 Cracker Blaster: drag pieces onto an 8×8 board, clear rows/columns, chain combos. The combo is deliberately forgiving and it survives a fumble, so it's kind to small hands. Vanilla JS, no build step. Audio is fully synthesised with the Web Audio API (chiptune tracks that crossfade as you heat up), plus a Three.js bloom backdrop. AI did the music and visuals 🤟 ▶️ https://cracker-blaster.vercel.app

• Post #3261475

  Working on TIL-Scope!! Tumour Microenvironment Atlas Single-cell report: the T-cell infiltrate An automated, reproducible scRNA-seq analysis of the tumour immune microenvironment, with an LLM-written interpretation aimed at a T-cell-engager discovery programme.

• Post #3261474

  Remove Background Noise from Video Without Re-encoding: An Audio-Only Approach with DeepFilterNet3 https://kemal.yaylali.uk/remove-background-noise-video-deepfilternet/ https://github.com/lynchaos/denoise