kagglehub

Complete Kaggle integration plugin/skill for AI coding agents — competition reports, dataset/model downloads, notebook execution, and badge collection. Works with Claude Code, Gemini CLI, Cursor, Codex, OpenClaw, and 35+ agents via skills.sh.

Deploy autonomous AI agents to research, debug, and iterate on Kaggle data science competitions.

Data analysis performed on my sleep logs

🧠 Deep learning project for brain tumor classification using MRI images. Built with transfer learning (VGG16 + fine-tuning), TensorFlow/Keras, and deployed via Streamlit. Dataset & model loaded dynamically from KaggleHub. Includes training notebook, evaluation, and interactive web app.

A deep learning research prototype designed to analyze real-world medical imaging and identify multiple thoracic pathologies using a fine-tuned DenseNet121 architecture, while also providing the exact regions that led to its predictions.

AI-powered plant disease detection system using CNN deep learning. Identifies 38+ crop diseases from leaf images with high accuracy. Built with TensorFlow/Keras and Streamlit for real-time agricultural diagnostics.

Python data cleaning pipelines that use Pandas and KaggleHub to download, clean, transform and prepare datasets for analysis and testing

Benchmarking Linear Regression, CatBoost, LightGBM, XGBoost, and an entity-embedding neural network on 1M vehicle listings

AI-powered pancreas detection and segmentation system for CT scans using CNN and U-Net models with automated mask generation and visualization.

AI Based Anomaly Detection System for System Logs

A content-based recommendation engine using LLMs, LangChain, Hugging Face, Gradio, and Python to generate vectorized book representations for semantic similarity matching, enabling personalized and context-aware book recommendations.

Pandas-based exploratory analysis of 1,025 heart disease patient records: KaggleHub loading, missing-value audit, iloc selection and boolean filtering in a reproducible Jupyter notebook.

a data analysis and forecasting project focused on identifying trends, top-performing brands, and year-over-year patterns in synthetic car sales data

This Machine Learning project presents a rigorous, leakage-aware hybrid soft-voting ensemble framework combining k-Nearest Neighbors (KNN), Support Vector Machine (SVM), and Random Forest (RF) for heart disease prediction on the UCI Cleveland dataset.

This repository is to experiment the integration with @ggml-org/whisper.cpp for offline STT + pyannote/speaker-diarization-3.1

Brain Tumor Classification

Exploring and Visualizing the Iris Dataset

This is project, developed a deep learning model using Tensorflow and CNNs to classify images into multi animal categories with high accuracy. Preprocessed image data, optimized model architecture and achieved effective generalization on unseen data.