search

Columbia-Basin-Research-CBR / SARforecastDLM

In development shiny app forecasting one-year wild SSSR Chinook salmon survival
0 stars 0 forks source link

readme

SARforecastDLM

Lifecycle: experimental

Last Commit

SARforecastDLM uses Dynamic Linear Modelling (DLM) to forecast changes in one-year ocean survival of wild spring/summer Snake River Chinook salmon, Oncorhynchus tshawytscha, from upwelling indices (CUI), West Coast, USA. This ShinyApp is developed based on the Scheuerell and Williams (2005).

Installation

Option 1: You can install the development version of SARforecastDLM from GitHub with:

# Install the developmental version of the HydroSurvSizePred package from GitHub
devtools::install_github("Columbia-Basin-Research-CBR/SARforecastDLM")

# Load the package
library(SARforecastDLM)

# Run the app
SARforecastDLM::run_app()

#if needed, detach the package workspace and repeat above lines of code
detach("package:SARforecastDLM", unload=TRUE)

SARforecastDLMis currently in development and changes are continuously being made. If you have already imported to R studio, please rerun install_github("Columbia-Basin-Research-CBR/SARforecastDLM") to see latest changes in the developmental version. If no changes have been made since last import, a warning will appear: Skipping install of 'SARforecastDLM' from a github remote, the SHAX (XXXXXX) has not changed since last install. Use 'force = TRUE' to force installation and you have the latest developmental version imported. Use detach() if not working properly and reinstall.

This option will run the Shiny App within you RStudio environment but will not download the background files necessary to run.

Option 2: If you are interested in the files that support the development version, please see: https://github.com/Columbia-Basin-Research-CBR/SARforecastDLM to clone the repository. Alternatively, within Rstudio, use the following steps:

  1. Start a new project with File > New Project > Version Control > Git

  2. In the repository URL field, paste https://github.com/Columbia-Basin-Research-CBR/SARforecastDLM.git

  3. Once project is created, and the repository is cloned, you can run the app within R environment by going to the folder: dev > run_dev.R and loading lines of code, with the final run_app() to launch the app.

Possible download error with GitHubs security settings:

If you receive an error associated with github_PAT credentials, you may need to setup a GitHub Personal Access Token (PAT) for authentication when you try to install a package from GitHub using devtools::install_github(). Here are the steps to resolve this issue:

  1. Generate a new GitHub PAT: Go to your GitHub account settings, then to Developer settings -> Personal access tokens -> Generate new token (classic or fine-grained). Set expiration date and save token. Token will disappear if not saved.

  2. Set the new PAT in your R environment: You can use the usethis package to set the PAT in your R environment. Here’s how:

# Install the usethis package if you haven't already
install.packages("usethis")

# Use usethis to set the GITHUB_PAT environment variable
usethis::edit_r_environ()

# This will open your .Renviron file in a text editor. Add the following line to the file, replacing "your_new_pat" with your actual PAT:
GITHUB_PAT=your_new_pat

# Save and close the file. Then, restart your R session to make sure the new environment variable takes effect.
  1. Try installing the package again: Now, you should be able to install the package from GitHub without encountering the authentication error. Follow steps in Option 1.

The app structure follows a Golem framework described in Engineering Production-Grade Shiny Apps by Colin Fay, Sébastien Rochette, Vincent Guyader and Cervan Girard.

Contact

This app is being developed by Caitlin O’Brien, Research Scientist, Columbia Basin Research, SAFS, University of Washington. Please reach out with questions/concerns via csobrien@uw.edu.