John Lu

1 month ago | nature.com | Jung-Ki Yoon |Yili Zhu |John Lu |Georgios Mikos |Camille Brenac |Nicholas J. Guardino | +6 more

AbstractFibrosis, the replacement of healthy tissue with collagen-rich matrix, can occur following injury in almost every organ1,2. Mouse lungs follow a stereotyped sequence of fibrogenesis-to-resolution after bleomycin injury3, and we reasoned that profiling post-injury histological stages could uncover pro-fibrotic versus anti-fibrotic features with functional value for human fibrosis. Here we quantified spatiotemporally resolved matrix transformations for integration with multi-omic data.

Aug 20, 2024 | lush93md.medium.com | John Lu

TL;DR: Recommendation system demand a deep understanding of not only the technical aspects but also the ability to interpret and apply evaluation metrics effectively. This blog post, inspired by foundational recommendation system concepts, will equip you with all the essentials you need. At the heart of many ML-driven applications is the recommendation system, an essential tool for enhancing user experience.

Jun 3, 2024 | lush93md.medium.com | John Lu

1. Generative Image Models and Latent Manifold Generative image models learn a "latent manifold" of the visual world. Think of this manifold as a low-dimensional vector space where each point corresponds to an image. When we have a point on this manifold, we can "decode" it to get a displayable image. The "decoder" model in the Stable Diffusion handles this decoding process. 2.

Jun 2, 2024 | lush93md.medium.com | John Lu

Stable Diffusion is like a digital artist that can turn words into pictures. It's a tool to create images from text descriptions. In this blog post, we're going to take a closer look at how this amazing tool works. Imagine telling a story and having Stable Diffusion draw the scenes - it's that kind of magic we'll uncover. 🎨✨With text-to-image generation, it's more effective to just show you how it works right from the start.

Apr 27, 2024 | lush93md.medium.com | John Lu

Fundamental Basics for Machine LearningI. Why do we need to normalize numerical features? II. The choice between precision and recallIII. How to tackle imbalance data? IV. Why can Dropout suppress overfitting? V. Briefly describe KNN(k-nearest neighbor) algorithm1. Equalizing ScalesNumerical features often have different scales (ranges of values). For example, consider age (ranging from 0 to 100) and income (ranging from 0 to 100,000+). Algorithms treat features with larger scales as more influential.