Quickly setup SSH connection to Kaggle Kernel for Deep Learning. In order to use that sexy Tesla P100 for free (and without many restriction of Jupyter Notebook) :P

Deep CNN-LSTM for Generating Image Descriptions 😈

All my Kaggle Notebooks that I've published

Kaggle's Photorealistic images of the Moon's surface with ground truth rock segmentation.

Notify new kaggle kernels to Slack or LINE without coding

A repo to Fine Tune BERT and use it for text classification.

A tool for kaggle kernel management.

🩺This project uses Random Forest classification to predict liver cirrhosis stages (1, 2, or 3) based on patient records from a Mayo Clinic study. By analyzing key clinical indicators such as bilirubin, albumin, and copper levels, the model supports early diagnosis and medical decision-making with interpretable machine learning.

🐾 This project builds a deep learning model to classify animals from images using transfer learning and CNNs. It processes visual data, predicts species with high accuracy, and presents results through an interactive dashboard—ideal for ecological research, education, and real-time applications.

💓This project applies Random Forest classification to predict the presence of heart disease using patient-level diagnostic and lifestyle features. By analyzing indicators like chest pain type, cholesterol, and ECG results, the model supports early diagnosis and risk stratification to assist healthcare professionals in making informed decisions.

A machine learning model that analyzes clinical and diagnostic features to predict the likelihood of thyroid cancer recurrence. It supports early intervention and personalized follow-up strategies for improved patient outcomes.

Detecting fraudulent financial transactions using a Random Forest Classifier trained on real-world transaction data. Includes .pkl file handling, feature selection, stratified sampling, and fraud probability prediction.

Machine Learning examples

Predicting lung cancer survival outcomes using clinical and lifestyle data. Includes preprocessing, feature engineering, and Random Forest classification with ~95% test accuracy. Built for healthcare risk stratification and early prognosis

This project builds a machine learning model to predict the price range of mobile devices based on their technical specifications. It uses preprocessing, visualization, and Random Forest classification to estimate cost categories ranging from low to very high.

Top 4%