Ahmed Hassan’s Post

🚀 Built a Machine Learning model that predicts house prices. Most people stay stuck in tutorials. I decided to apply it. Used Linear Regression to train on real housing data, evaluated performance, and saved the model for reuse. 📊 Results: • R² Score: 0.58 • MSE: 0.56 Not perfect, but real learning happens here building, testing, improving. Pushed the complete project to GitHub 💻 #BuildInPublic #MachineLearning #AIJourney #Python #DataScience #Consistency #KeepLearning

🚀 Hands-on Machine Learning Project: Decision Tree Classifier Recently, I worked on a small but insightful project where I implemented a Decision Tree Classifier using Python and Scikit-learn. 📊 What I did: Created a structured dataset with features like Age, Salary, and Experience Applied data preprocessing techniques Built and trained a Decision Tree model Evaluated performance using Confusion Matrix & Classification Report Visualized patterns using Seaborn 📈 Key Learnings: How Decision Trees split data based on feature importance Importance of handling data properly before modeling Understanding evaluation metrics like precision, recall, and F1-score 💡 This project helped me strengthen my fundamentals in machine learning and model evaluation. 🔗 I’ll be sharing the GitHub repository soon! #MachineLearning #DataScience #Python #ScikitLearn #DecisionTree #DataAnalytics #LearningJourney

We are taking the training wheels off. 🚲 In Part 7, we used the "Easy Button" to build an AI agent. Today, in Part 8, we are opening up a Jupyter Notebook and building a custom RAG pipeline from absolute scratch using Python. If you want to move from "Full-Stack Developer" to "Data Scientist / AI Architect," you have to understand the math beneath the magic. In this tutorial we cover: 🔪 Programmatic Text Chunking 🔢 Generating Vector Embeddings (text-embedding-004) 📐 Calculating Cosine Similarity with numpy to build a semantic search engine. Read the full tutorial here: https://lnkd.in/ewtWxBT6 #Python #DataScience #MachineLearning #VertexAI #GoogleCloud #VectorSearch

🚗 Car Price Prediction using Machine Learning Happy to share my ML project where I built a model to predict car prices based on various features. 🔧 Technologies Used: - Python - Scikit-learn - Pandas, NumPy 📌 Key Features: ✔ Data preprocessing ✔ Model training & evaluation ✔ Prediction system 🔗 GitHub Repository: https://lnkd.in/g_yrducF 🎥 Project Demo: [Paste your video link here] #MachineLearning #Python #DataScience #CodeAlpha #AI

Today, I focused on working with NumPy arrays. Building a solid foundation for data manipulation and analysis. Here’s what I practiced: 🔹 Created a 1D array with values from 1 to 15 🔹 Built a 2D array (3×4) filled with ones 🔹 Generated a 3×3 identity matrix 🔹 Explored key array properties like shape, type, and dimensions 🔹 Converted a regular Python list into a NumPy array This session helped me better understand how data is structured and handled in numerical computing. Getting comfortable with arrays is definitely a crucial step toward more advanced data analysis and machine learning tasks. Looking forward to building on this momentum 💡 #AI #MachineLearning #Python #NumPy #DataAnalysis #M4ACE

Day 6 of Solving ML Problems From Scratch: Adam Optimizer Today I worked on implementing the Adam Optimizer from scratch. What I like about Adam is that it combines the benefits of momentum and adaptive learning rates in a very practical way. Instead of taking the same type of step every time, it adjusts based on both past gradients and gradient magnitude, which makes optimization more stable and efficient. While solving this, I got a better understanding of: how momentum helps smooth the update direction how the velocity term adapts the step size why bias correction is important, especially in the early steps how Adam can converge faster than plain SGD in many cases Building these concepts from scratch is helping me understand what is really happening behind the libraries we use every day. It is one thing to call an optimizer in code, but it is very different to actually implement and reason through each update step yourself. Small daily practice like this is making machine learning feel much more intuitive. #MachineLearning #DeepLearning #ArtificialIntelligence #Python #DataScience

Day 72. Spent time going deeper into XGBoost today. Covered classification and worked through the math: gradients & hessian leaf weights similarity score & gain Some questions I tried to answer while learning: Why do we need Taylor expansion here? Why can’t we directly differentiate the objective? What makes decision trees non-smooth / non-differentiable? The key realization: since trees produce piecewise constant outputs, the loss surface isn’t smooth — which is why second-order approximation becomes necessary. Still revising, but things are starting to connect. Notes: https://lnkd.in/gCqHUeK9 #MachineLearning #XGBoost #LearningInPublic #Python #DataScience

Excited to share my latest project: LinearRegression-ML This is a beginner-friendly Machine Learning project focused on understanding and implementing Linear Regression from scratch. It includes practical notebooks like profit analysis and medical data predictions, along with clear explanations of loss and cost functions. ???What I learned =>Fundamentals of Linear Regression =>Cost & loss function implementation =>Real-world dataset analysis using Python #https://lnkd.in/guCQQdNe #MachineLearning #Python_Jupyter_Notebook #DataScience

Seaborn vs Matplotlib — what’s the difference? While learning data visualization, I explored both libraries and here’s my simple understanding - 📊 Matplotlib 🔹 Basic and highly customizable 🔹 More control over plots 🔹 Requires more code 📊 Seaborn 🔹 Built on top of Matplotlib 🔹 More visually appealing 🔹 Easier to use for statistical plots 💡 My takeaway: Matplotlib gives control, Seaborn gives simplicity and better visuals. Using both together is the best approach. Which one do you prefer? #Python #Seaborn #Matplotlib #DataVisualization #LearningInPublic

I trained a model last week. Got 97% accuracy. Showed it to the team. Everyone was thrilled. Then I changed one line of code. The accuracy dropped to 61%. Same model. Same data. Same algorithm. The only thing that changed? The random seed. I've spent the last few weeks building something to explain exactly why this happens — and how to make sure it never happens to you. Stay tuned. Spoiler: Had I used #skore library from :probabl. along with scikit-learn, this would have never happened.... https://lnkd.in/eMmwpj8a #skore #datascience #mlops #dataanalytics #crossvalidation #python #sklearn