Star Cluster Simulation Analysis Project

USU cs5040/6040 group 10

Marcus Quincy, Reagan Baxter, Eli Peterson

Data and Results

This project is to create visualizations from the Star Cluster Simulation Data. It is a set of 19 csv files.

Some of the main goals of the project are to analyze which stars escape and also gain a better visualization of the sparse data set.

The main tools used for this are ParaView (https://www.paraview.org/) and python.

Description of files in this repo

• animations/

  • Directory containing various video animations of the data.

• csvs/

  • data/
    • Directory containing the original data (with .bak extension), the and the results of running the preprocess script both with separating the escapees out, and with standard preprocessing. This directory can be used as input to escapeesColor.pvsm.
  • escaped-interpolated/
    • Directory containing the csvs which are the output of running the cubic-interpolation script. This directory can be in input to escapee-cubic-interpolated.pvsm.
  • splitData/
    • Directory containing data split into two groups: one with ID and one without. The data here is input to escapeeSpeed.pvsm and EscapeesTemporalInterpolated.pvsm.
      • The set of csvs named ids_c_*.csv contain a speed column that is used in escapeeSpeed.pvsm and plotting.py

• plots/

  • distance_plot.png
    • a png of the distance plot generated by plotting.py
  • speed_plot.png
    • a png of the speed plot generated by plotting.py

• stateFiles/

  • escapee-cubic-interpolated.pvsm
    • ParaView state file for creating an animation of the escaped stars with cubic interpolation. Expects data to be preprocessed by cubic-interpolation script first.
  • subset-cubic-interpolation-animated.pvsm
    • ParaView state file for creating an animation of random stars with cubic interpolation. Expects data to be preprocessed by cubic-interpolation script first.
  • escapeesColor.pvsm
    • ParaView state file coloring escaped stars green through all the time steps. Requires escapees to be separated using preprocess script first.
  • escapeeSpeed.pvsm
    • ParaView state file which analyzes the speeds of the stars that escape. Uses the ids_c_*.csv data set.
  • EscapeesTemporalInterpolated.pvsm
    • State file with just the stars that escape and uses Temporal Interpolation to visualize their movements. Requires data to be processed by Split.py first.
  • temporalInterpolator.pvsm
    • State file with temporal interpolation applied to the entire data set. Uses the base data set.
  • interpolatedPathlines.pvsm
    • State file containing the animation of the escaping stars pathlines. Uses the escaped-interpolated set of csvs.

• python/

  • Centroid.py
    • Uses data processed by fixMissingIds.py or preprocess as input. This script calculates the mean of all the data for visualization.
  • cubic-interpolation
    • Takes the output of preprocess script as input. This script takes the data and outputs a data set with more time steps with cubic interpolation applied.
  • fixMissingIds.py
    • Adds id's to the data to compensate for all the stars missing data.
  • plotting.py
    • Generates 2 plots. One of the speeds of the stars in relation to the time steps and the other of the stars distance from center over the time steps. Uses the ids_c_*.csv csvs found in splitData.
  • Split.py
    • Splits the data into escapees and nonescapees.

Steps to reproduce

Requirements

• ParaView (5.11 required for cubic interpolation animation state file)
• python3
• numpy
• scipy
• pandas
• seaborn
• matplotlib

All these can be installed via guix shell -m manifest.scm.

Pre Process the data

Note: most of the processed data has been saved already in the repository.

Cubic Interpolation

First missing ids (escapees) need to be populated back into the csv files. This is done with the preprocess script. (If cloned this repo this step is done for you...)

./python/preprocess -i.bak --separate-escapees csvs/data/c_*.csv

This will also copy the data for the ids that escape into another set of csv files. The output of these files will be in the same directory that the files were originally in.

For cubic interpolation run the following. This will create ~700MB of data for the entire csv set. For example to generate cubic interpolation data for the escaped stars with a time step every 0.02 time steps of the original run:

./python/cubic-interpolation -n 0.02 -o <output dir> <path to csv e.g. csvs/data/escapee_c_*>

Centroids

For Centroid calculation, first edit the input and output files in the source file to the desired location then run python python/Centroid.py. The tail of the file should look like:

if __name__ == "__main__":
    input_dir = "csvs/splitData"  # Change this to the input directory containing CSV files
    output_dir = "my-output-dir"  # Specify the output directory to save centroid CSV files

    process_csv_files(input_dir, output_dir)

Plots

To view the plots analyzing the data:

python3 python/plotting.py

This will use matplotlib and seaborn to visualize the data. Close the first plot to see the second.

Load the Paraview state files

Now with the data processed, the Paraview state files can be loaded. Be sure to select the data manually instead of using the paths in the state file. Which CSV files are expected to be associated with each ParaView file is described in the "Description of files in this repo" section above (note not all CSV files are part of the repo to prevent it from getting bigger than it already is).