A collection of pandas & scikit-learn compatible transformers for preprocessing and feature engineering 🛠

Optimizing an ML Pipeline in Azure - A Machine Learning Engineer Project

Data Science case for windmill power prediction based on weather. Based on Data Challenge of Air Liquide (@AirLiquide) and TotalEnergies(@total-sa, @Total-RD) companies in 2021. Real Data Science model was much more complicated to get 6 place. The link of the competition - https://datascience.total.com/fr/challenge/19/details#.

[Template] To Create MCPStack MCP Tools in ~5 command lines!

Scikit-Learn useful pre-defined Pipelines Hub

Programming assignments covering fundamentals of machine learning and deep learning. These were completed as part of the Plaksha Tech Leaders Fellowship program.

A Modular, Production-Style ML Pipeline with Class-Imbalance Handling

Analysis on effect of temperature, rainfall and pesticide on Global food crops yield along with Crop Yield prediction using scikit-learn ML algorithms. Interactive website for outcomes using HTML, Bootstrap, JS, CSS, d3 and so on.

Build custom vacab, Ham /Spam using tfidf , Movie review classification using TFIDF

Code repository for "MyeVAE: A multi-omics variational autoencoder for predicting mortality in newly diagnosed myeloma patients"

An interactive HR-analytics app predicting employee churn via scikit-learn pipelines and Streamlit

This Repo contains the Mini-Projects I've made while self-learning Data Science.

Unsupervised anomaly detection on 3-year multivariate sensor data from a cyclone preheater using Python. Detected 437 abnormal timestamps using Isolation Forest, One-Class SVM, and Z-score filtering. Includes visual timeline, anomaly impact summary, and Excel highlights.

Laptop Cost Evaluator A fast, user-friendly web tool that predicts the market price of a laptop based on key specs like brand, processor, RAM, and storage. Just enter the details on a sleek, single-page form, and get an instant price estimate thanks to a smart backend powered by machine learning.

Multi-omics integration for the classification of functional high risk patients in multiple myeloma.

Student dropout predictions based on grades and other info. Classification problem with MLPClassifier.

A Jupyter notebook project to predict customer conversions in digital marketing using Random Forest (0.97 accuracy). Includes EDA, preprocessing, model building, evaluation, and Streamlit deployment to enhance campaign targeting, boost conversion rates, and maximize ROAS.

A Machinehack ML challenge