===================== Data Science BootCamp ===================== Python Packages =============== Pandas Data Frames Numpy Numberical Python Scipy Scientific Python Sklearn Neural Network library Matplotlib Matlab style plotting library Plotly Interactive plotting library Lasagne Neural Networks in Python Data Acquisition ================ Data Loading ------------ Data Exploration ================ Data Visulations ---------------- Data Modelling ============== Linear Model ------------ Linear Regression ***************** k-nearest neighbours -------------------- Radial Neighbors kNearest Neighbors Computationaly expensive Data Data needs to be normalised Performance Evaluation Decision Trees -------------- Decision Trees: Prone to overfitting Works with Categorical Works with unnormalised data Based on asking questions Cost is logarthimic Depth of tree is main issue