Skip to main content

Working with Anduryl

Prerequisites​

You can either install these programs separately or install Anaconda.

Anduryl Installation​

  1. Plan where you want Anduryl installed.
    • We suggest you install the module somewhere easily accessible like your Home folder.
    • Terminology used:
      • PARENT_FOLDER: This is the path to the folder that you want Anduryl installed into.
        • This can be anywhere on your system, it will look similar to this (you can change this)
          Windows: C:/Users/YOUR_NAME
          Mac: /home/name
  2. Download Anduryl. This can be done 2 ways, downloading it from your browser, or using the git command.
    • Option 1: Browser
      1. Navigate to https://github.com/GBADsInformatics/anduryl-plus
      2. Click on the green Code button then Download ZIP
      3. Move the downloaded ZIP to the PARENT_FOLDER you would like Anduryl installed in.
      4. Right click the ZIP and click Extract All or Extract Here
      5. Rename the new anduryl-plus-master folder to anduryl
      6. Open your operating systems Command Prompt or Terminal
      7. Run cd INSERT_PARENT_FOLDER_HERE
      8. Run python -m anduryl
    • Option 2: Git Command (requires git to be installed)
      1. Open your command prompt/terminal, then navigate to your PARENT_FOLDER.
        • Run cd INSERT_PARENT_FOLDER_HERE
      2. Run git clone https://github.com/GBADsInformatics/anduryl-plus.git
      3. Rename anduryl-plus to anduryl
        • To do so, run mv anduryl-plus anduryl
      4. Run python -m anduryl

Anduryl in Jupyter Notebook​

Import Anduryl

  • To import Anduryl, Jupyter needs to know where to find it. Add the block of code below to the top of your Jupyter Notebook to import the module.
    Note: Change the code to your PARENT_FOLDER before you run it.
#Showing the system where to find Anduryl
import sys
anduryl_path = 'INSERT_PARENT_FOLDER_HERE' # Path to anduryl parent folder
if anduryl_path not in sys.path: sys.path.append(anduryl_path) # Add path to sys
import anduryl

Usage Examples

Anduryl in R Studio​

  1. Install the reticulate library to call python modules from R.
    • In R, type install.packages("reticulate")
    • Follow the wizard to install reticulate
  2. Import reticulate and Anduryl by running the code below.
    Note: Change the first line to your PARENT_FOLDER
    anduryl_path = 'INSERT_PARENT_FOLDER_HERE'
    library("reticulate") 
    sys <- import("sys")
    if (!anduryl_path %in% sys$path)(sys$path <- append(sys$path, anduryl_path))
    anduryl <- import("anduryl")
  3. Each time you load up R, you'll need to re-run step 2.
  4. Usage example:
    proj <- anduryl$Project()
    proj
    This code should output 
    Main project class.
    Properties:
     - assessments
     - experts
     - io
     - items
     - main_results
     - results
    Methods:
     - add_results_from_settings
     - calculate_decision_maker
     - calculate_expert_robustness
     - calculate_item_robustness
     - initialize
     - to_results
  • To learn more about coding with Anduryl, visit Example Anduryl with scripting
  • Note: because you will be coding in R, you will reference objects using $ instead of .
    • Ex. anduryl$Project() instead of anduryl.Project()