GALAssify: A Python package for visually classifying astronomical objects
Bibliographic record
Abstract
Summary The visual classification of astronomical objects requires the use of tools that are simple and easily adaptable to the requirements of the user. In this context, we present GALAssify, a graphical tool that allows the user to visually inspect and characterise properties of astronomical objects in a simple way. In addition, GALAssify allows the user to save the results of the visual classification into a file using a list of previously defined tags based on the user's interests. GALAssify is available on GitLab and on PyPI. For many classification problems faced in astrophysics, a graphical interface greatly facilitates the job. In the present work we focus on the classification of galaxies to present a software that helps in a customised and simple way to do this classification. GALAssify has been developed in Python using PyQt5 libraries. A priori, it has been initially developed to tackle astrophysical problems but, due to its versatility, it could be easily adapted. For instance, this tool can be used to classify microscopy images from biological studies or be used in any other discipline. We provide instructions for the installation, usage and basic examples of how to use GALAssify. In the GALAssify's GitLab we show the use of GALAssify for visual classification of galaxy morphology. However, the user can extrapolate this visual study to any image. Statement of need Python is currently one of the most widely used programming languages in the scientific community, particularly in astrophysics. We have developed GALAssify with the aim of facilitating scientists and collaborators the task of visually classify the desired properties of astronomical objects. Additionally, the results can be easily shared between collaborators for analysis and comparisons and can be used for scientific reporting. GALAssify interface was initially designed to perform the galaxy sample selection in the CAVITY (Calar Alto Void Integral-field Treasury surveY) project. CAVITY is a survey aimed to study galaxies in voids using Integral Field Unit data (Pérez et al. 2024). Usage GALAssify is a tool to classify images from a list of user-defined tags. To do so, we provide a list of galaxies, the equatorial coordinates right ascension and declination (ra and dec, respectively), the path to the figures, and the relevant tags for the classification. There are three types of buttons that can be selected, radiobutton (only one of the options in the list can be selected), checkbox (the desired number of options can be selected) and comments. This particular usage has been widely used and tested within the CAVITY collaboration. The left panel of the GUI shows the list of galaxies, which is a table with the following columns: assignation, the name of the galaxy, an icon indicating whether it has been processed, and the coordinates ra and dec for each galaxy. In the upper part of the right panel, the image of the selected galaxy is displayed. Optionally, this panel can also display the corresponding FITS image of the selected galaxy, specified by the user. In the case of galaxies observed in the SDSS, the algorithm allows the user to provide a path to the figure (if the image is located on their computer) or download it from the SDSS or DESI websites given its coordinates. The lower part of the right panel is divided into three sections, where we show the different classifying options. The classification of each galaxy can be edited or reset at any time. Finally, to save the selection, one can simply click the "Save and next button" or press the "enter" key. The entire classification is saved in a comma-separated values (CSV) file, easily readable with any text editor, spreadsheet program or database manager. For a detailed usage and configuration guide, please visit the official documentation hosted on GitLab Pages. We also provide additional support tools to: Download images from SDSS and DESI/DECaLS catalogues. Create an instructions pdf document to guide the user through the graphical user interface (GUI). Documentation Package documentation is available on the GALAssify's' GitLab Pages. Software Citations GALAssify should work with Python >= v3.9 and makes use of the following packages: PyQt5 Pandas NumPy Matplotlib Requests Pillow Astropy pyds9 PyConsoleMenu Only in case the user wants to display the images in fits format, it is necessary to have SAOImageDS9 installed on the system. License The code is licensed under MIT License. Acknowledgements We acknowledge financial support by the research projects AYA2017-84897-P, PID2020-113689GB-I00, and PID2020-114414GB-I00, financed by MCIN/AEI/10.13039/501100011033, the project A-FQM-510-UGR20 financed from FEDER/Junta de Andalucía-Consejería de Transforamción Económica, Industria, Conocimiento y Universidades/Proyecto and by the grants P20_00334 and FQM108, financed by the Junta de Andalucía (Spain). M.A-F. acknowledges support from the Emergia program (EMERGIA20_38888) from Junta de Andalucía. G.B-C acknowledges financial support from grants PID2020-114461GB-I00 and CEX2021-001131-S, funded by MCIN/AEI/10.13039/501100011033, from Junta de Andalucía (Spain) grant P20-00880 (FEDER, EU) and from grant PRE2018-086111 funded by MCIN/AEI/10.13039/501100011033 and by 'ESF Investing in your future'. SDP acknowledges financial support from Juan de la Cierva Formación fellowship (FJC2021-047523-I) financed by MCIN/AEI/10.13039/501100011033 and by the European Union `NextGenerationEU'/PRTR, Ministerio de Economía y Competitividad under grants PID2019-107408GB-C44 and PID2020-113689GB-I00, from Junta de Andalucía Excellence Project P18-FR-2664, and SDP is grateful to the Natural Sciences and Engineering Research Council of Canada, the Fonds de Recherche du Québec, and the Canada Foundation for Innovation for funding. TRL acknowledges support from Juan de la Cierva fellowship (IJC2020-043742-I), financed by MCIN/AEI/10.13039/501100011033.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.006 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.003 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.004 |
| Open science | 0.004 | 0.005 |
| Research integrity | 0.001 | 0.003 |
| Insufficient payload (model declined to judge) | 0.069 | 0.053 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".