Shahzaib Hamid’s Post

Module 4 at #llmzoomcamp shows the fundamental principles of evaluating LLM, this provide an insight on creating evaluation metrics to accurately measure the performance of LLM in your application. Check the course repo: https://lnkd.in/eZk_Qung

RAGFlow, the deep document understanding based #RAG engine is open sourced, offering a streamlined RAG workflow for businesses, combining #LLM to provide truthful question-answering with solid citations from data of complex formats. 👉 https://lnkd.in/gak6YJDe

Acronyms

Exciting news from TEAM IM! We’re thrilled to announce that we are in the process of creating a brand-new redaction service, combining the best features of our previous offerings into a single, powerful application. This innovative service allows efficient document rendering in a browser and seamless processing with integrated redaction API endpoints. Users can define specific text strings or regular expressions to find and hide sensitive data, view and manage redactions directly on the document using the our viewer technology, and easily review all redactions listed on the side of the page. Learn more here: https://bit.ly/4dOY2vB #PrizmDoc #RedactionService #DataSecurity #DocumentManagement #TEAMIM #TechInnovation #DataProtection #SeamlessRedaction #PrivacyFirst #SecureDocuments #TEAM #insights #teaminsights

🤔 Guess The Word Challenge! 🌐 Can you unravel this term that signifies a connected system in both professional and technological realms? Drop your guesses in the comments! #wordchallenge #guesstheterm #nerdyservice #nerdyservicept #nerdyservicenl #digitalservices #nerdychallenge #guesstheword

🚀 Dive into the performance world of ImmutableArray<> in this insightful article! 👩💻 Compare the SequenceEqual() method with a conventional for() loop and discover which one triumphs in benchmarks. 📊💡 Check it out: https://lnkd.in/gvGpXY3Q #dotnet8 #MVPBuzz #CodePerformance

So today I finally found out what exactly RAG is thanks to #llmzoomcamp. In case you're wondering yourself, it's a fancy term to say: "give an LLM (chatGPT) some notes about a specific question you have, and tell it to find the answer within those notes". That's it. That's literally all it means. Oh and in case you want a less ELI5 explanation, check out the course: https://lnkd.in/dpyMS5ks

This is week 2 of #mlzoomcamp and it was all about Linear Regression. You can view my homework here: https://lnkd.in/dV3wpe-8 The goal of this homework was to create a regression model for predicting laptop prices (column 'Final Price') and I did that in the following steps: 1️⃣ EDA and preparing the dataset by splitting it into training, validation and testing set. 2️⃣ Shuffling and normalizing the data by applying log transformation. 3️⃣ Building a linear regression from scratch using Numpy. 4️⃣ Evaluating a model on a validation set by using the Root Mean Squared Error (RMSE). 5️⃣ Making predictions with the model. The focus of the lessons for this week was learning how to perform Linear Regression from scratch, without using any machine learning library, and it was great to know how the 🚗 Case study: Car price prediction 🤖 Data Cleaning & Exploration 🧮 Linear Regression 🔄 Root Mean Squared Error 🐍 Feature Engineering 📚 Regularization 📊 Model training and using the model and others. PS: I know I said I'm excited about github codespaces in week 1, but I now take it back. It requires stable internet, the kind that MTN Nigeria is unable to deliver on, so I'm back to VSCode and GitBash to preserve my sanity. #mlzoomcamp #datascience #machinelearning #machinelearningzoomcamp #IT #womanintech #data #learninginpublic #growth #pushthrough #upskilling #thisIHarderThanIThought #EnjoyingIt

Check out our article about remote access scams and learn how to stop them. 🛡 👇 #AnyDeskScam #TeamViewerScam #TechSupportScam https://lnkd.in/df2J7xym

I am super interested in developer facing information resources (docs sites) that have AI chat as the primary starting point to access and explore the information. Pinecone's docs at https://lnkd.in/gtFR6jG6 is a good example of this. If you know of another good example, please share! Have you taken it a step further and gone totally in on an AI chat experience indexed on a flat corpus instead of curating IA (Information Architecture) as a means of discovery ? If so, please share! And what pros, cons did you experience?