suppressPackageStartupMessages(library(AER)) suppressPackageStartupMessages(library(tidyverse)) suppressPackageStartupMessages(library(broom)) suppressPackageStartupMessages(library(nlme)) suppressPackageStartupMessages(library(lme4)) suppressPackageStartupMessages(library(lmerTest)) Introduction Basic regression models are fitted with a sample of \(n\) independent elements. Given a set of \(p\) regressors \(X_{i,j}\) and a continuous response \(Y_i\), we fit a model

Introduction This is a personal Rmarkdown document I have created to visualize the COVID-19 updates and some preliminary exploratory data analysis (EDA). The source of this data is the github repository created and maintained by the Coronavirus COVID-19 Global Cases by the Center for Systems Science and Engineering (CSSE) at Johns Hopkins University (JHU).

Public morale was low in BC during the peak of the COVID-19 pandemic in November 2020, as a second wave of infections was imminent Two substantial social distancing interventions were implemented within two weeks of each other to curve rising numbers of COVID-19 prior to the holidays: - November 7th: Restricted private gatherings between households - November 19th: Mandatory masks all public spaces, and non-essential travel was heavily discouraged What if these measures were not implemented at that time? What if they had been implemented sooner? How would daily case numbers have be impacted in these scenarios? We explored these possibilities using a compartmental SEIR model in R, focusing on October 1st to December 15th

Recent developments in deep learning algorithms such as generative adversarial networks (GANs) (Goodfellow et. al. 2014) has created a paradigm shift in architectural generative design. Advantages of GANs over traditional generative algorithms include the ability to generate and manipulate infinite number of unique surfaces. Training of such models include feeding a large library of surfaces into the network, the model then learns the important patterns and features of the training data to generate realistic examples. The model can be divided into a generator and a discriminator. The generator objective is to learn how to generate representations of the input training examples that can “deceive” the discriminator. The discriminator is a universal function approximator that outputs a probability of an input surface to be real or fake. As more and more training epochs are passed, a minimax loss function is minimized that allows for the generator to create surfaces that are close as possible to the real ones, and the discriminator to not be able to differentiate between real and fake images.