/Kaggle_challenges

1) Pima_Indians_Diabetes_dataset: 83.11% accuracy. 2) Using ScanPy for gene expressions in scRNA sequencing. 3) Using traditional clustering on RNA dataset 4) Stroke Prediction 95% accuracy on test

Primary LanguageJupyter Notebook

Kaggle_challenges

  • Pima Indians Diabetes dataset - classification models. Random Forests: 83.11%, XGBoost: 82.6%, GradientBoost: 81.16%, LightGBM: 79.87%, Neural Network: 75.32%
  • Stanford Ribonanza RNA Folding dataset - Similarity scoring
  • Clustering of scRNA based on gene expression similarity using K-Means, Leiden, Louvain and UMAP techniques