ivv101 / cidview

Multidimensional visualization

Settings panel screenshot

About

This tool allows for easy interactive visualization of a csv file with multidimensional numerical data.

When applicable, one can specify a feature as a class feature so that the the objects belonging to each class will have a unique color, could be shown/hidden etc.

The user can choose from multiple features by clicking the corresponding buttons to the left of the Y-axis and below the X-axis. At the top there are buttons corresponding to the distinct classes. Users can choose to display different combinations of the classes by toggling/untoggling the class names. Whenever appropriate, a logarithmic scale is used in the plot (which is reflected in the axis labels). The user can select a source by clicking on it and some information about the source will be shown. A lasso tool can be used to select a group of sources which then remain selected for any combination of features plotted. The plot can be easily saved as a PNG file by clicking the corresponding button to the right of the plot.

CIDview has benn used to visualize data for several astronomical datasets:

• XCLASS -- visualizes the content of the multiwavelength training dataset (Yang et al. 2021) based on Chandra Source Catalog.

• XDBS -- visualizes the content of the X-ray detected Be star catalog (Gobat et al. 2022).

• XCLASS_XMM -- visualizes the content of the multiwavelength training dataset (Yang et al., in preparation) based on 4XMM X-ray source catalog.

• GCLASS -- visualizes the content of the 4FGL-DR4 Fermi Gamma-rag Observatory LAT catalog.

Installation

1. Clone this repo and

cd CIDview

2. [optional] Create environment

mamba create -n CIDview
mamba activate CIDview

2. Install dependencies

mamba env update --file environment.yml

3. Test installation:

python code/CIDview.py data/gll_pcs_v28.csv settings/gll_pcs_v28_settings.json

The code will generate index.html in html folder.

Usage

python code/CIDview.py <csv_file> [settings_file.json]

The Settings panel (top left corner) allows the user to customize most of the GUI. The settings can be saved and then used to recreate the html file.