Last week, Ellis scraped F1 data as a first step in answering a viewer question about how to create your own R package. This week, we take the data that we scraped and begin the initial stages of building an R package. We discuss the steps in setting up an R package, creating the description file, adding the scraping code from last week to a data-raw folder, and then building the data into the data-export folder.
To watch the screen cast, CLICK HERE.
To access our code, CLICK HERE.
Ellis Hughes and I have been building up an interface for the user to input a data set and then run a regression model and get out of sample predictions. This week, add more to our {shiny} model user interface and allow the user to specify the type of model they want to run. Instead of just allowing a linear regression model the user can now choose to either run a random forest or a decision tree.
To watch our screen cast, CLICK HERE.
To access our code, CLICK HERE.
In this next iteration of our Shiny Modeling building series, Ellis Hughes and I advance our user defined shiny model to ensure that the user is not able to select a variable as both a predictor and outcome variable. We discuss two different ways to do this, based on what sort of controls you’d like to provide the user. Additionally, we cover how to add a Toast drop down box to the user interface. When the user changes something, for example builds a model and then decides they want to change the outcome variable, a little drop down will pop out at the top of the screen to inform the user that something has changed with their data. This ensures that the user is always aware of the output they are looking at, helping to prevent any mistakes in interpretation.
To watch our screen cast, CLICK HERE.
To access our code, CLICK HERE.
This week, Ellis Hughes and I extend our Shiny Modeling Building episode from LAST WEEK and provide the user an opportunity to decide how much of their data to allocate to a training data set. We go through how to build a {shiny} server that can take the user inputs, separate a reactive training and test set based on those inputs, build a regression model, and make out of sample predictions.
To watch our screen cast, CLICK HERE.
To access our code, CLICK HERE.
This week, Ellis Hughes and I build a {shiny} app that allows users to upload a .csv file, select a dependent variable and any number of independent variables and then get returned results about Mean Absolute Error, Residuals, and Predictions.
To watch our screen cast, CLICK HERE.
To access our code, CLICK HERE.
