Partnering with Dev teams to improve cloud native DevEx | Business Development @ mirrord, by MetalBear 🐻 | ex-BCG
mirrord is simpler to use than to describe, so Anatoly Tarnavsky crafted a clever analogy to shed light on the experience of using it. What do you think? Big thanks to Anatoly Tarnavsky for the sharp summary!
Java Developer |HackerRank 3🌟 coder|Inovations | frontend developer @CSE 27'|Student at SATI Vidisha M.P
Day10 solved successfully... 🚀 🚀 Sample Problem summary: to find maximum sum of subarray and return the sum. Solving approach: 💥 💥 By using kadane's Algorithm. . assume res = arr[0]; and maxEnding=arr[0]; . Iterate loop for i=i to i=arr.size() .find maximum subarray maxEnding=max(maxEnding+arr[i],arr[i]; . finally return sum 💻 💻 Complexities Through this algorithm.. . Time Complexity: O(n) . Auxiliary Space: O(1) #gfg160 #geekstreak2024
Prisma is a multimodal mechanistic interpretability library based on TransformerLens, currently supporting vanilla vision transformers (ViTs) and their vision-text counterparts CLIP. The goal of Prisma is to make research in mechanistic interpretability for multimodal models.
Laying the Foundations for Vision and Multimodal Mechanistic Interpretability & Open Problems by Sonia Joseph, Neel Nanda Outline Part One: Introduction and Motivation Part Two: Tutorial Notebooks Part Three: Brief ViT Overview Part Four: Demo of Prisma’s Functionality Key features, including logit attribution, attention head visualization, and activation patching. Preliminary research results obtained using Prisma, including emergent segmentation maps and canonical attention heads. Part Five: FAQ, including Key Differences between Vision and Language Mechanistic Interpretability Part Six: Getting Started with Vision Mechanistic Interpretability Part Seven: How to Get Involved Part Eight: Open Problems in Vision Mechanistic Interpretability 👉 https://lnkd.in/drg49E5F #machinelearning
Software Engineer | Deep Learning | Machine Learning and AI | Software Design | Student at Brigham Young University - Idaho
Such a good course if you are trying to understand quantization for LLM's!
Definition of: 'mellifluous' is: 'sweetly or smoothly flowing; sweet-sounding.'. Learn more at: 'https://lnkd.in/dAvyGbbP'
In this episode, we discuss DepthCrafter: Generating Consistent Long Depth Sequences for Open-world Videos by Wenbo Hu, Xiangjun Gao, Xiaoyu Li, Sijie Zhao, Xiaodong Cun, Yong Zhang, Long Quan, Ying Shan. DepthCrafter is a novel method for estimating temporally consistent depth in open-world videos without needing additional data like camera poses or optical flow. It generalizes to diverse video content by utilizing a three-stage training strategy rooted in a pre-trained image-to-video diffusion model, enabling it to handle up to 110-frame sequences. Evaluations show DepthCrafter's state-of-the-art performance, bolstering applications like depth-based visual effects and conditional video generation.
Infosearch's Annotation Services: Data annotations - Image, Video, Text, Audio annotations etc are provided through various techniques like bounding box, polygon, cuboid, semantic segmentation and more. Learn more about our annotation services here: https://lnkd.in/gWujs2H4 Contact us to outsource your services: https://lnkd.in/gWujs2H4 enquiries(@)infosearchbpo(.)com #annotation #annotationServices #dataAnnotation #imageAnnotation #boundingBox #polygons
I've been diving deep into algorithmic problem-solving and recently tackled some classic challenges like the 2-sum, 3-sum, and 4-sum problems! 🎯 Revision is the key.... Especially, ARRAY is one of the topic which includes --> Index-Based Access --> Understanding Array Bounds --> n-Place Modifications --> Pattern Recognition**** --> Edge Cases
How can you traverse a binary tree to record all paths through it? BinaryTreeTraversal #RecordPaths #CodingProjects #TreeAlgorithms Understanding the Need for Path Traversal in a Binary Tree 🌳 Are you working on a project that involves navigating through a binary tree to capture every distinct path from the root to the leaves? You're not alone! This kind of problem can be tricky because standard tree traversal techniques like inorder, preorder, or postorder aren... Source: https://lnkd.in/g3kfgRd5 #mymetric360
Faculty @ Zhejiang University; Adjunct Faculty @ Monash University; Visiting Research Scientist @ DAMO Academy
We'd like to share our latest simple and effective work, ZipAR, which reduces 91% of auto-regressive image generation overhead without any training. 🔮 🔮 Addressing the issue of slow decoding in AR image generation models (such as Emu3, Lumina-mGPT, LlamaGen, and Janus), we propose a training-free parallel decoding framework that achieves up to 91% reduction in image generation overhead while preserving almost all image details and semantic information. This is an early-stage work, and we are continuing to improve it. Feedback and discussions are very welcome! Paper link: https://lnkd.in/gizCB4bK Code link: https://lnkd.in/gjDvmEsh
Have you heard about Domain Experiments? On Friday I had the pleasure of talking to Mike Wojtyna about Domain Experiments. It is an alternative to design level event storming. But not only. I think it is a great approach for rapid prototyping combined with domain discovery. It is easy to validate different models against unit tests. Another great tool to add to your #dddesign toolbox. Watch the full conversation with Mike Wojtyna: https://lnkd.in/dvifrv63 What do you think?
