Mastering User Interaction in Python with Satish Dhawale

🎬 Built a Movie Recommendation System using Python & ML Using content-based filtering to recommend movies based on similarity — here's a quick breakdown how it works: 1. TF-IDF Vectorizer converts movie descriptions into vectors. 2. Cosine Similarity measures how similar two movies are. 3. Random Forest classifier validates the results. Results: ◈ 16 movies ◈ 4 genres ◈ 97% model accuracy ◈ Toy Story & Finding Nemo topped similarity at 0.61 ◈ The Godfather & Goodfellas closely matched at 0.58 A great way to understand how Netflix-style recommendations work under the hood. Open to feedback and questions! 👇 #MachineLearning #Python #DataScience #RecommendationSystem #BuildInPublic #dataanalytics #datascience #mlproject #datanalyst #datascientist #scikitlearn

You can fit the most common Bayesian regression models in Python using a consistent syntax (similar to brms in R) using the bambi package. It utilizes PyMC to do the simulations. It's remarkably easy and straightforward to use - you just adjust the family name to the right model type. Here are a few examples. More instructions are available here: https://lnkd.in/eGSG3-Bk #statistics #datascience #analytics #rstats #python #peopleanalytics #technology #ai

🐍 Day 89 — Features and Labels Day 89 of #python365ai 📌 Features (X) → input variables Labels (y) → output Example: X = [size, rooms] y = price 📌 Why this matters: Clear distinction is essential for building ML models. 📘 Practice task: Identify features and labels in a dataset. #python365ai #Features #MachineLearning #Python

Learn in Public – Day 13 Today I practiced implementing the Armstrong Number problem in Python using two different approaches. 🔹 Approach 1: Converting the number to a string and calculating the power of each digit. 🔹 Approach 2: A more mathematical approach using functions like order() and a recursive power() function. Key learnings from today: • How Armstrong numbers work mathematically • Practicing recursion through a custom power function • Working with digits using modulus and integer division • Writing cleaner modular functions Example: 153 → 1³ + 5³ + 3³ = 153 Consistency > Motivation. #LearnInPublic #Python #CodingPractice #ProblemSolving #100DaysOfCode

This problem demonstrates how dictionary comprehensions can create key–value pairs dynamically. It highlights how values can be generated from existing data while building a dictionary in a compact and readable way. A very useful technique when transforming datasets. THE ANSWER IS: B #Python #DictionaryComprehension #PythonChallenge #DataStructures #BuildInPublic

Wrapped up an AI project for our coursework recently — probably the most hands-on thing we've built so far. The stack: Python, LangChain, HuggingFace embeddings, Chroma, and a few LLM tools. Getting all of that to actually work together was its own learning curve, but seeing it click into place was worth it. Classroom theory only takes you so far. Actually building something — debugging it, breaking it, fixing it — is a different experience entirely. Glad we got to do both. More to come. #ArtificialIntelligence #MachineLearning #AIProjects #TeamProject #Python

Day 6 of #LearnInPublic Today I worked on the problem: First Non-Repeating Element in an Array. I implemented two approaches in Python: 1️⃣ Hash Map Approach (Using defaultdict) • Count frequency of each element • Traverse again to find the first element with frequency = 1 Time Complexity: O(n) Space Complexity: O(n) 2️⃣ Brute Force Approach • Compare every element with the rest of the array Time Complexity: O(n²) Space Complexity: O(1) Key takeaway: Using a hash map trades extra space for a significant improvement in time complexity. Small daily improvements compound over time. #Python #DataStructures #Algorithms #LearnInPublic #CodingJourney

Built my own RAG (Retrieval-Augmented Generation) system from scratch using Python Worked on the complete pipeline: Extracting text from PDF Splitting into meaningful chunks Creating embeddings using Ollama Storing data in Chroma vector database Retrieving relevant information using similarity search Next step is to integrate an LLM and turn this into a full AI-powered application 💡 #AI #MachineLearning #Python #RAG #Ollama #BuildInPublic

Before working on the AI employee, spent time learning the core: 𝘀𝘂𝗯𝗽𝗿𝗼𝗰𝗲𝘀𝘀 𝗶𝗻 𝗣𝘆𝘁𝗵𝗼𝗻. Understanding how Python can execute external commands, capture stdout and stderr, control execution with timeouts, and work with clean string outputs instead of raw bytes. Also looked into managing return codes and controlling how external tools run from inside a Python program. Small piece, but it’s the bridge between Python and the outside world. #Python #Subprocess #Learning

Built Logistic Regression from Scratch using NumPy! Implemented the sigmoid function, trained the model using gradient descent, and visualized the logistic curve for binary classification. This project helped me understand how logistic regression actually works under the hood without using ML libraries. 🔗 GitHub: https://lnkd.in/gnZ-g4aQ #MachineLearning #Python #NumPy #LogisticRegression #DataScience #LearningInPublic