Statistical Learning and Deep Learning from NTUIM department.
The goal of this course is to introduce a set of tools for data analytics. This course cover the principles and applications of these models. These tools will not be viewed as black boxes. Instead, we will explore these models' details, not just the use. The main reason is that no single approach will perform well in all possible applications.
In each folder, the problems will be _Q, and my solutions be _Sol.
The score for Hw4 isn't good, but the rest are almost 100.
K-means regressor, hyperparameter tuning, Lasso Regression, Ridge Regression
Data preprocessing, 1-hot-vector, ROC and AUC curve, Logistic Regression with L2 regularization
Logistic Regression, Random Forest, Gradient Boosting Decision Tree, Stacking, Data Visualization via Dimensionality Reduction
Exploring Multilayer Perceptrons for Regression, Minibatch, Epoch, and Early Stopping
Kaggle competition, training CNN model to categorize 4 different kind of outfit, got kaggle score around 0.95-0.96.
Prediting possible diabete patients with multiple health conditions, such as BMI, blood pressure... and so on.