Pinar Yalçın Güler’s Post

I am very happy and proud to share that, together with my brilliant students and colleagues Pinar Yalçın Güler, Muhammed Enis Sen and Buğra Eyidoğan, we have established HDMRLib, a tensor decomposition library for Python. This project reflects our collective research efforts and aims to provide practical, scalable implementations for high-dimensional modeling and tensor-based methods. We will continue to improve the library and expand it with new features actively. Please feel free to use it, and we would greatly appreciate your thoughts, feedback, and suggestions. GitHub: https://lnkd.in/dJU6yvVS Documentation: https://lnkd.in/d5gqQMwe

Built a Machine Learning API using FastAPI I developed a machine learning-based API that predicts salary based on user input level. My all project and machine learning model based API on github. GitHub : https://lnkd.in/gR_qsxwM 🔹 Implemented Machine Learning algorithms and integrated them with FastAPI 🔹 Enabled real-time prediction using API based on user input 🔹 Designed RESTful endpoints for seamless interaction 🔹 Stored and retrieved prediction data dynamically 💡 This project demonstrates how ML models can be deployed and used through APIs in real-world applications. Tech Stack: Python, FastAPI, scikit-learn #MachineLearning #FastAPI #Python #DataScience #AI #BackendDevelopment #MLProjects

🚀 Finished my GenAI assignment on **Prompt Templates using LangChain** I made a simple "Mini Prompt Engine" that turns user input into structured prompts using templates that can be used again instead of hardcoded strings. 💡 What I did: ✅ Used PromptTemplate to make prompts that change over time ✅ Made prompts with multiple inputs (topic, audience, tone) ✅ Made prompts in different styles (teaching, interview, storytelling) ✅ Added basic input validation ✅ Tested to see if the template could be used again 🛠 Tech Stack: Python, LangChain, and Jupyter Notebook 📌 Learned how to make prompt systems that can be used in real-world AI applications and are flexible. # GitHub: https://lnkd.in/gCBdN3Jw GenAI #LangChain #PromptEngineering #Python #AI #InnomaticsInnomatics Research Labs

Most people jump directly into Machine Learning models. I almost did the same. But then I realized something: Without strong fundamentals, everything in ML becomes confusing. So instead of rushing into algorithms, I’m currently focusing on: • Data Structures & Algorithms (for problem-solving) • Probability & Statistics (to actually understand models) • Python fundamentals (clean implementation matters) Because in the long run: Understanding why something works is more powerful than just knowing how to use it. Now I’m building my learning step by step — and documenting it along the way. Curious to know — how did you approach learning ML? #DataScience #MachineLearning #Python #DSA #LearningInPublic

Wrapped a session of the Harvard AI / Python course today and it sharpened a few things for me. What stood out: • Python is less about syntax and more about thinking clearly. Break problems down properly and the code follows. • AI models are only as good as the data and assumptions behind them. That responsibility sits with us. • The real power is in building small working pieces fast, then stacking them into something useful. • It’s practical, buildable, and ready to deploy into real workflows. I’m already thinking about how this feeds directly into Mana Review AI — tighter models, cleaner data pipelines, better decision support. This is the level-up phase. #AI #Python #GovTech #IndigenousTech #Harvard

𝗧𝗶𝗺𝗲 𝘀𝗲𝗿𝗶𝗲𝘀 𝗳𝗼𝗿𝗲𝗰𝗮𝘀𝘁𝗶𝗻𝗴 𝗶𝘀 𝗼𝗻𝗲 𝗼𝗳 𝘁𝗵𝗼𝘀𝗲 𝘁𝗼𝗽𝗶𝗰𝘀 𝘁𝗵𝗮𝘁 𝗹𝗼𝗼𝗸𝘀 𝘀𝗶𝗺𝗽𝗹𝗲... 𝘂𝗻𝘁𝗶𝗹 𝘆𝗼𝘂 𝘁𝗿𝘆 𝘁𝗼 𝗱𝗼 𝗶𝘁 𝗽𝗿𝗼𝗽𝗲𝗿𝗹𝘆. Most tutorials stop at fitting ARIMA or Prophet on a clean dataset. In practice, it’s messier: – missing data – leakage – feature engineering – evaluation pitfalls – deployment constraints That’s where most real-world projects fail. There’s an upcoming workshop that focuses exactly on this gap — end-to-end forecasting in Python, not just models in isolation. What I like about it: – covers the full pipeline (data → features → models → evaluation) – hands-on, not just theory – focused on practical decisions you actually face If you're working with time series (or planning to), this is a useful one to check. 👉 https://lnkd.in/dnergVrT Use code 𝗔𝗡𝗗𝗥𝗘𝗬𝟰𝟬 for 40% off. Event date: May 2 #MachineLearning #TimeSeries #DataScience #Forecasting #MLOps

I built research_copilot, an AI-powered tool designed to simplify the research workflow. This tool can: • Process research papers (PDFs) • Extract key insights and summaries • Generate knowledge graphs between concepts • Identify research gaps • Assist in drafting literature reviews It is built using React and a Python (Flask) backend, integrated with an AI reasoning engine for handling long-context academic analysis. This project has enhanced my understanding of backend system design, including handling file uploads, structuring data processing pipelines, and integrating external APIs. Sharing a quick demo below: GitHub: https://lnkd.in/dT_7vffa #AI #BackendDevelopment #Python #Projects #Learning

The Art of Focus: Mastering Image Cropping with NumPy! 🎯✂️ Day 86/100 In a world of data noise, the ability to focus on what matters is a superpower. For Day 86 of my #100DaysOfCode journey, I explored Region of Interest (ROI) Extraction. In Computer Vision, we don't always need the full picture. By using NumPy Array Slicing, I can 'zoom in' on specific coordinates to isolate faces, text, or objects for further analysis. Technical Highlights: 🎯 ROI Identification: Mastering the coordinate system to pinpoint and extract sub-matrices from large image arrays. ✂️ Precision Slicing: Leveraging Python's [start:stop] syntax to perform lossless cropping in microseconds. ⚡ Computational Optimization: Learning why reducing image size via cropping is the first step in high-speed object detection. 🤖 AI Preprocessing: Understanding how cropping helps prepare datasets for deep learning models by removing irrelevant background noise. Do check my GitHub repository here : https://lnkd.in/d9Yi9ZsC #100DaysOfCode #ComputerVision #NumPy #Python #BTech #IILM #AIML #ImageProcessing #DataScience #SoftwareEngineering #LearningInPublic #WomenInTech

🚀 Why Python is the Backbone of Data & AI (My Practical Understanding) Most beginners learn Python as just a programming language. But in reality, Python is a complete problem-solving ecosystem. 💡 Here’s how I see it (my practical understanding): ✔ Data Analysis → Pandas ✔ Numerical Computing → NumPy ✔ Data Visualization → Matplotlib / Seaborn ✔ Machine Learning → Scikit-learn ✔ AI / Deep Learning → TensorFlow, PyTorch ⚙️ What makes Python powerful? • Simple and readable syntax → faster development • Multi-paradigm support → flexible problem-solving • Massive library ecosystem → ready-to-use solutions 🔍 Technical Insight (Important): Python is not just an interpreted language. It first converts code into bytecode, which is then executed by the Python Virtual Machine (PVM) — making it platform-independent. #Python #DataAnalytics #AI #MachineLearning #CareerGrowth #TechSkills