Hunters, busybodies, and the knowledge network building associated with curiosity

Code for the paper, Hunters, busybodies, and the knowledge network building associated with curiosity.

Setup

platform       x86_64-pc-linux-gnu         
arch           x86_64                      
os             linux-gnu                   
system         x86_64, linux-gnu           

R:             3.5.1 (2018-07-02)
MATLAB:        9.6.0.1174912 (R2019a)
Python:        2.7.15

See environment_root.yml for Python libraries and packages used.

Author

Dale Zhou (dalezhou [at] pennmedicine.upenn.edu)

Project Organization

    ├── data                                    <- Data goes here.
    │      |
    │      ├── subjectLevel
    │      ├── kFolds
    │
    ├── scripts                                 <- Downloaded functions go here
    │      |
    │      ├── copyScripts.sh                   <- prepare code to fit each individual
    │      ├── editScripts.sh                   <- prepare code to fit each individual
    │      ├── entropySimulated.py              <- function for entropy
    │      ├── errwLevyFunction.py              <- function for growth model
    │      ├── errwLevyFunction.m               <- MATLAB version of growth model
    │      ├── heapsSimulated.py                <- function for Heaps' law
    │      ├── intervalsSimulated.py            <- function for inter-event time
    │      ├── launchAnalysis.sh                <- launch training on cluster
    │      ├── launchTest.sh                    <- launch test on cluster
    │      ├── nsga.py                          <- main script running evolutionary optimization
    │      ├── testFit.py                       <- main script testing fit
    │      ├── wikiWrangler.R                   <- prepare code to fit each individual
    │      ├── zipfsSimulated.py                <- function for Zipf's law
    │
    │
    ├── environment_root.yml                    <- Python environment packages
    │
    ├── README.md

Order of scripts

1. Run wikiWrangler.R to get train and test folds
2. copyScripts.sh if copyCommands does not exist. Then source copyCommands
3. editScripts.sh if editCommands does not exist. Then source editCommands
4. qsub launchAnalysis.sh to launch the nsga.py scripts
5. qsub launchTest.sh to launch the testFit.py scripts

Notes

The growth model itself is errwLevyFunction.py or errwLevyFunction.m for equivalent Python and MATLAB versions. All other code is to fit individual data to that growth model.

Scripts were run on a high-performance computing cluster.