Gravity Spy: Glitch classification and dataset
Annotating Gravitational Wave Bursts
By [source]
Paper: Machine learning for Gravity Spy: Glitch classification and dataset. Bahaadini, et al.
About this dataset
The Gravity Spy Training Set is a crowd-sourced collection of labelled glitch waveform images, used to train a convolutional neural network for the Gravity Spy project. With this data set, citizents scientists are able to classify and annotate gravitational wave bursts in order to further understand the mysteries of space!
This dataset contains numerous columns containing metadata which users can utilize in order to analyze and make sense of gravitational waves. These columns include event_time, ifo, peak_time, peak_time_ns, start_time, start_time_ns, duration ,search ,peak frequency ,central frequency ,channel ,amplitude ,snr chisq & chidof as well as parameter one name & value. All of these fields serve an important purpose for classifying and analyzing signals from space!
Whether you’re an experienced scientist or a curious enthusiast seeking answers about the universe ) - this dataset is ideal for anyone who wants to uncover some gravitation secrets! So come explore our archive today and help unlock some cosmic mysteries
More Datasets
For more datasets, click here.
Featured Notebooks
- 🚨 Your notebook can be here! 🚨!
How to use the dataset
In order to use this dataset, we recommend that you first read through the column descriptions provided above. Each column provides information about a particular aspect of each waveform image, such as its event time, peak time/frequency, channel used to detect it, amplitude/signal-to-noise ratio (SNR), etc. There are also columns that provide URLs for up to 4 related images associated with each waveform image available in the training set.
After familiarizing yourself with the contents of this dataset and its features/labels in detail, you can now dive into your own analysis and exploration of gravitational waves! You may want to filter out events according to certain criteria (e.g. channel used), look at specific images more closely or compare different aspects between them (e.g SNRs). Additionally this data can be used alongside other datasets from sources such as LIGO open science center or Gravitational Wave Open Science Center in order to gain further insights into these mysterious cosmic activities!
We hope that you thoroughly enjoy exploring this incredible universe of knowledge with us - happy kaggling!
Research Ideas
- Training machine learning models to detect and classify gravitational wave signals in real-time.
- Developing algorithms to accurately measure the amplitude and frequency of gravitational waves in a given time frame.
- Creating interactive visualizations or animations that showcase the various effects that different gravitational wave frequencies can have on our universe
Acknowledgements
If you use this dataset in your research, please credit the original authors.
Data Source
License
License: CC0 1.0 Universal (CC0 1.0) - Public Domain Dedication
No Copyright - You can copy, modify, distribute and perform the work, even for commercial purposes, all without asking permission. See Other Information.
Columns
File: trainingset_v1d1_metadata.csv
| Column name |
Description |
| event_time |
The time of the event in UTC. (String) |
| ifo |
The interferometer used to detect the event. (String) |
| peak_time |
The time of the peak of the event. (Float) |
| peak_time_ns |
The time of the peak of the event in nanoseconds. (Integer) |
| start_time |
The start time of the event. (Float) |
| start_time_ns |
The start time of the event in nanoseconds. (Integer) |
| duration |
The duration of the event. (Float) |
| search |
The search algorithm used to detect the event. (String) |
| peak_frequency |
The frequency of the peak of the event. (Float) |
| central_freq |
The central frequency of the event. (Float) |
| channel |
The channel used to detect the event. (String) |
| amplitude |
The amplitude of the event. (Float) |
| snr |
The signal-to-noise ratio of the event. (Float) |
| chisq |
The chi-squared value of the event. (Float) |
| chisq_dof |
The degrees of freedom of the chi-squared value. (Integer) |
| param_one_name |
The name of the first parameter associated with the event. (String) |
| param_one_value |
The value of the first parameter associated with the event. (Float) |
| label |
The label assigned to the event. (String) |
| sample_type |
The type of sample used to detect the event. (String) |
| url1 |
The URL of the waveform image associated with the event. (String) |
| url2 |
The URL of the second waveform image associated with the event. (String) |
| url3 |
The URL of the third waveform image associated with the event. (String) |
| url4 |
The URL of the fourth waveform image associated with the event. (String) |
Acknowledgements
If you use this dataset in your research, please credit the original authors.
If you use this dataset in your research, please credit .