Pith. sign in

REVIEW 4 major objections 7 minor 53 references

Capturing star formation activity from compressed photometric images of galaxies

T0 review · 4 major / 7 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read A Vision Transformer trained on 124,000 compressed survey color images classifies star-forming galaxies with F1=0.85 and predicts emission-line ratios with R^2≈0.83, suggesting spectroscopy can be bypassed for the star-formation question.

desk verdict A clean, honest image-only star-formation classifier that is missing the color-only control needed to support its central claim about visual features. read the letter →

arxiv 2507.23019 v1 pith:HU4XCJGN submitted 2025-07-30 astro-ph.GA

Kyuseok Oh , M. Dennis Turp This is my paper
classification astro-ph.GA
keywords AstronomicalmethodsNeuralnetworksAstronomydataanalysisimageprocessingstar-forminggalaxiesVisionTransformerBPTdiagramemission-lineratios
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper claims that the star-forming status of a galaxy can be read directly from ordinary compressed survey images, without spectra. Using a Vision Transformer on about 124,000 optical color-composite JPEGs of nearby galaxies at $0.01

What carries the argument

The load-bearing mechanism is the Vision Transformer base model with 16×16 patches on 224×224 images: each image is split into patches, embedded into a 768-dimensional space, and summarized by a [CLS] token through 12 self-attention blocks, so local clumpy star-forming regions and global spiral structure both enter the prediction. Pre-training on natural images is transferred to galaxy images, and task-specific heads turn the [CLS] representation into either two class logits or two continuous line-ratio outputs. Self-attention over patches is what lets the model combine color and spatial pattern without hand-designed morphology features.

What would settle it

Take the same train/test split and fit a classifier using only the $g-r$ color (or any single color index) of each galaxy image; if a color-only baseline reaches F1≈0.85 and $R^2\approx0.83$, the claim that visual morphology carries the star-forming signal is falsified.

Watch

Extended reading notes

Core claim

On its own terms, the paper's discovery is that a galaxy's spectroscopic ionization state leaves a recoverable imprint in its optical appearance. A Vision Transformer base model, fine-tuned on the g, r, i color composites, classifies star-forming versus non-star-forming galaxies with precision 0.85, recall 0.86, and F1 0.85; for the galaxies it labels star-forming, it predicts $\log([{\rm N\,II}]λ6583/{\rm H}\alpha)$ and $\log([{\rm O\,III}]λ5007/{\rm H}\beta)$ with $R^2 = 0.837$ and $0.830$, tracing the star-forming ridge of the BPT diagram with reduced scatter. The authors also report that the mapping does not transfer to $z\approx0.16$-$0.17$ and that nuclear-dominated AGN ratios are not recoverable, because the nuclear emission occupies less than a pixel in the compressed images.

Load-bearing premise

The predictive signal must come from spatial or visual features, not just from the integrated color information already encoded in the JPEG; the paper tests no simple $g-r$ color baseline, so if such a cut matches F1≈0.85, the direct-link claim collapses to a color statement.

Editorial extensions

If this is right

  • If the claim holds, star-forming galaxy catalogs for future wide surveys can be produced from survey images alone, turning a spectroscopy-limited question into an imaging pipeline.
  • The regression result implies that BPT-style line ratios of star-forming galaxies can be ranked or binned from images, enabling statistical studies of ionization and metallicity without individual spectra.
  • The model's sharp drop at $z\approx0.16$-$0.17$ defines the usable redshift range: the visual-to-spectroscopic mapping works only where galaxy structure is resolved in more than a few pixels.
  • The reported accuracy applies only to the coarse two-way split; the four-way BPT classes (star-forming, composite, Seyfert, LINER) are not recovered from JPEGs.
  • Because false negatives are biased toward massive star-forming galaxies, downstream science using this classifier must account for a mass-dependent incompleteness.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A color-only shortcut is the most serious untested alternative: a simple $g-r$ (or similar color-index) threshold could plausibly reproduce much of the F1, since star-forming galaxies are predominantly blue and JPEG composites encode color; the paper provides no such baseline, so its visual-features interpretation remains open.
  • A direct way to isolate what the network actually uses is to feed grayscale versions of the same images or to mask the central pixel region; if performance persists, the model relies on light distribution and morphology rather than on color alone.
  • The same architecture could act as an anomaly detector: galaxies whose predicted emission-line ratios disagree strongly with measured ones are likely interacting, edge-on, or hosting unresolved nuclear activity, exactly the cases the paper shows are misclassified.
  • The higher-redshift failure mode points to resolution rather than intrinsic physics, so deep, high-resolution imaging from upcoming surveys may recover the mapping if angular size is the limiting factor.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 7 minor

Summary. The paper trains a Vision Transformer (ViT) on approximately 124,000 SDSS gri color-composite JPEG images of galaxies at 0.01<z<0.06 to perform two tasks: (1) binary classification of galaxies into star-forming versus non-star-forming (with non-star-forming defined as the union of composite, Seyfert, LINER, weak-emission, and non-emission galaxies), and (2) regression of the [NII]λ6583/Hα and [OIII]λ5007/Hβ line ratios. The spectroscopic labels are taken from the OSSY catalog. The authors report a test-set F1 score of 0.85 for classification and R² values of 0.837 and 0.830 for the two line-ratio regressions. They further evaluate the model on unseen redshift bins (0.06<z<0.07 and 0.16<z<0.17), finding comparable performance at the lower bin and a marked degradation (F1=0.56) at the higher bin. The paper interprets these results as evidence that photometric JPEG images alone can directly capture star-forming activity, bypassing traditional spectroscopic diagnostics.

Significance. If the central claim is robust, the method would provide a cheap, fast screening tool for star-forming galaxies in large photometric surveys, which is a practically useful contribution. The paper is honest about limitations: it reports the high-redshift degradation, the failure for Seyferts and LINERs, and the lossy nature of JPEG compression. It also provides code and data through a public repository, which is commendable. The external OSSY labels make the supervised mapping non-circular. However, the significance of the claimed 'direct link' between visual features and spectroscopic state is currently underevidenced, because the paper never rules out that the model is simply exploiting the well-known color–BPT correlation (e.g., blue star-forming versus red passive galaxies). The novelty and interpretation therefore rest on a missing control experiment.

major comments (4)
  1. [Section 4.1, Section 6] The central claim of a 'direct link between apparent visual features and underlying spectroscopic characteristics' (Section 6) requires demonstrating that the ViT captures spatial/morphological information beyond integrated color. The paper never compares against a simple color-based baseline, such as a g-r color cut or a logistic regression on SDSS g-r, u-r, and r-i colors (or on the average RGB of the same JPEG images). Since star-forming galaxies are systematically bluer (the paper itself cites Strateva et al. 2001), a color-only model could plausibly match F1=0.85, especially given the coarse binary grouping that lumps all non-star-forming types together. Please add such baselines and report F1, precision, recall, and regression R² for them. If the ViT does not significantly outperform the color-only model, the abstract's claim that images 'directly' encode star-forming activity collapses to a restatement of known color–BPT correlations.
  2. [Section 4.2, Figure 5] The regression quality is only reported via R², RMSE, and MAE, but the authors themselves state in Section 4.2 that predictions are 'skewed toward the peak values' and that the predicted BPT diagram shows 'smaller dispersion' than the ground truth. These statements indicate strong shrinkage toward the mean, which can inflate R² when the test labels are concentrated near the mode. Please quote the linear-fit slopes and intercepts (currently only in the figure) and compare the RMSE (0.089 and 0.161) with the standard deviation or 16–84th percentile width of the ground-truth line-ratio distributions. This is needed to assess whether the regression predicts anything beyond the median of the training population.
  3. [Section 4.3, Section 6] The abstract and Section 6 state that the method is promising for Euclid, DES, and LSST, but the extrapolation to 0.16<z<0.17 yields F1=0.56, which the authors attribute to reduced angular size and loss of resolved features. Since LSST and Euclid will predominantly deliver galaxies at z>0.1, the claimed applicability is not supported by the reported experiments. Either the survey-applicability statements should be substantially tempered, or the authors should provide evidence of a transferable variant (e.g., training on higher-redshift galaxies, using larger cutouts, or leveraging multi-band coadds with better resolution). As written, the high-redshift degradation undermines the stated practical motivation.
  4. [Section 2, Appendix C] The binary classification collapses composites, Seyferts, LINERs, weak-emission, and non-emission galaxies into a single 'non-star-forming' class. The confusion matrix in Figure 4 does not break down performance by subtype, yet composites are known to be intermediate in color and often classified as star-forming by color-based methods. Please provide a subtype-level confusion matrix (e.g., precision/recall for star-forming versus composite versus passive galaxies) to clarify whether the model is separating 'star-forming' from 'everything else' by color or by actual star-formation indicators. This is directly relevant to the baseline concern raised above.
minor comments (7)
  1. [Section 4.2] Please state explicitly that all emission-line ratios are logarithmic (log10) before reporting numerical values such as RMSE=0.089; the axes of Figure 5 are not labeled in the text, and the numerical errors are ambiguous without this unit convention.
  2. [Section 1] The first paragraph contains a typo: 'Sloand Digital Sky Survey' should be 'Sloan Digital Sky Survey'.
  3. [Section 5] The heading 'LIMITATIONS AND CA VEATS' should read 'LIMITATIONS AND CAVEATS'.
  4. [Section 4.3] The phrase 'c.f.,' should be 'cf.,' for the Latin abbreviation, and the sentence 'when the extract features are insufficient' in Section 4.2 should read 'when the extracted features are insufficient'.
  5. [Section 3.2.3] The authors state that the best model is selected based on RMSE (regression) or accuracy (classification) on checkpoints saved every 100 steps, but they do not report the final epoch or step at which the best model was obtained; please provide this information for reproducibility.
  6. [Section 2] The text says a lower redshift limit is imposed but does not specify it in Section 2; the abstract states z=0.01, so please state the lower limit explicitly in the data section.
  7. [Section B, Figure 9] The discussion of morphology notes that 'uncertain' objects dominate all groups (>50%) and have small apparent sizes, but the text does not explicitly caution that the Galaxy Zoo morphological interpretation is therefore limited to a minority of relatively large, clearly classified objects; please add such a caveat.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the ViT is trained on external spectroscopic labels and evaluated on held-out data, so the results do not reduce to the paper's own inputs.

full rationale

The paper's derivation chain is a standard supervised learning pipeline: it takes SDSS gri composite JPEG images as inputs and trains a Vision Transformer against spectroscopic labels and line ratios from the OSSY catalog (Oh et al. 2011) and the follow-up measurements of Oh et al. (2015). The labels are generated externally, using Gaussian amplitude-to-noise thresholds and BPT demarcation lines from Kewley et al. (2001), Kauffmann et al. (2003b), Kewley et al. (2006), and Schawinski et al. (2007), not from the image model itself. The reported F1 and R^2 values are evaluated on a held-out 20% test split (Section 3.2.3) and on disjoint redshift bins (Section 4.3), so the predictions are not refits of parameters derived from the same data. The only self-referential element is that the first author co-authored the OSSY catalog, but that catalog is used as an external ground-truth resource with published line measurements and does not presuppose the visual-features-to-spectroscopy claim. The absence of a color-only baseline is a legitimate scientific weakness about what information the model uses, but it is not a circularity: no equation in the paper defines the image representation in terms of the BPT classes or line ratios, and no fitted quantity is renamed as a prediction. Thus no circular step can be exhibited, and the appropriate score is 0.

Assumptions & free parameters 1 free parameters · 4 assumptions · 0 invented entities

No new physical entities are introduced. The free parameters are minimal: the only hand-chosen numeric threshold is the A/N>=3 cut that defines the emission-line sample. The main assumptions are data-quality and representation assumptions, listed above.

free parameters (1)
  • A/N>=3 threshold for emission-line classification = 3
    Hand-chosen signal-to-noise cutoff for labeling emission-line galaxies in OSSY; changing this threshold changes the class definitions and thus the target labels. Section 2.
assumptions (4)
  • domain assumption OSSY catalog emission-line measurements are accurate enough to serve as ground truth.
    The model is trained and evaluated against OSSY (Oh et al. 2011), which is itself derived from SDSS spectra with stellar population models. Any systematic errors in OSSY become the model's target. Section 2.
  • domain assumption JPEG color composite images retain sufficient information to infer BPT line ratios.
    The entire method depends on this. The authors themselves note in Section 5 that JPEG discards flux precision and can lose nuclear emission, which weakens the assumption for AGN-dominated galaxies.
  • ad hoc to paper The SDSS gri color composite mapping preserves the physical color information correlated with BPT activity.
    The exact stretch and color assignment used to produce the JPEGs are not specified, yet the model's success relies on this fixed mapping. Section 3.
  • domain assumption The random 80/20 split does not create train-test overlap that inflates performance.
    No deduplication or grouping by SDSS plate is described; galaxies with similar photometry may appear in both sets. Section 3.2.3.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Capturing star formation activity from compressed photometric images of galaxies." pith.science (2026). https://pith.science/paper/HU4XCJGN

@misc{pith2026250723019,
  author       = {Pith},
  title        = {Pith review of: Capturing star formation activity from compressed photometric images of galaxies},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/HU4XCJGN}},
  note         = {Machine review of arXiv:2507.23019}
}
abstract

We present a novel approach for classifying star-forming galaxies using photometric images. By utilizing approximately $124,000$ optical color composite images and spectroscopic data of nearby galaxies at $0.01<z<0.06$ from the Sloan Digital Sky Survey, along with follow-up spectroscopic line measurements from the OSSY catalog, and leveraging the Vision Transformer machine-learning technique, we demonstrate that galaxy images in JPEG format alone can be directly used to determine whether star-forming activity dominates the galaxy, bypassing traditional spectroscopic analyses such as emission-line diagnostic diagrams. We anticipate that this method holds significant potential for application in current and future large-scale surveys, such as Euclid, the Dark Energy Survey (DES), and the Legacy Survey of Space and Time (LSST).

Figures

Figures reproduced from arXiv: 2507.23019 by the authors.

Figure 1
Figure 1. Pie chart describing the parent samples (left panel) and the [N ii]λ6583/Hα diagnostic diagram (right panel). Left panel: Outline of the parent samples, divided based on emission-line characteristics. The loci of star-forming galaxies, composite galaxies, LINERs, and Seyferts, which comprise 57.3% of the parent samples, are displayed in the right panel. Right panel: Emission-line classification of the SDSS data usin… view at source ↗
Figure 2
Figure 2. Flowchart illustrating a general overview of this work. The Vision Transformer (ViT), shown in the middle with orange diamonds (Section 3), is fed SDSS gri color composite images and used to predict emission line ratios. Section 4 describes the results in detail [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Vision Transformer architecture for galaxy classification and regression. mation that projects this representation to two output logits, corresponding to the two classes. The model is trained using cross-entropy loss, and during inference, a softmax function transforms these logits into class probabilities, where the probability for the star-forming class indicates the model’s confidence in the galaxy be￾ing star-fo… view at source ↗
Figures from the paper (8 more)
Figure 4
Figure 4. Figure 4: Confusion matrix illustrating the results of the binary classification between star-forming galaxies and non￾star-forming galaxies on the held-out 20% test set. The num￾ber of sources in each category is displayed, with the corre￾sponding fraction in parentheses and ca…
Figure 5
Figure 5. Figure 5: Upper-panels: Comparison of emission-line ratios (predictions versus ground truth). The test dataset, selected from the star-forming galaxies (NSF = 11,879), is used. The upper-left and upper-right panels show the [N ii]λ6583/Hα and [O iii]λ5007/Hβ line ratios, respect…
Figure 6
Figure 6. Figure 6: Confusion matrix illustrating the results of the binary classification between star-forming galaxies and non-star￾forming galaxies, applied to the evaluation dataset in the redshift range 0.06 < z < 0.07 (left panel) and 0.16 < z < 0.17 (right panel). The format is con…
Figure 7
Figure 7. Figure 7: Upper-panels: Comparison of emission-line ratios (predictions versus ground truth) for the test set. The upper-left and upper-right panels show the [N ii]λ6583/Hα and [O iii]λ5007/Hβ line ratios, respectively. Predicted emission-line ratios for star-forming galaxies ar…
Figure 8
Figure 8. Figure 8: Distribution of the angular diameter of galaxies, inferred from the SDSS isophotal major axis in the r band (isoAr), in units of arcseconds (left panel), and the star formation rate (SFR, in M⊙yr−1 ) as a function of stellar mass (right panel). Left panel: The four cat…
Figure 9
Figure 9. Figure 9: Left panel: Morphology classifications from the Galaxy Zoo 1 data release (Lintott et al. 2011), grouped according to each category in the confusion matrix presented in [PITH_FULL_IMAGE:figures/full_fig_p015_9.png]
Figure 10
Figure 10. Figure 10: Upper-panels: Comparison of emission-line ratios (predictions versus ground truth). The test dataset, selected from the emission-line galaxies (Nemi. = 14,232), is used. The upper-left and upper-right panels represent the [N ii]λ6583/Hα and [O iii]λ5007/Hβ line ratios…
Figure 11
Figure 11. Figure 11: Error distributions for the [N ii]λ6583/Hα (left panel) and [O iii]λ5007/Hβ (right panel) emission-line ratios. The dotted and dashed lines indicate the RMSE and MAE, respectively [PITH_FULL_IMAGE:figures/full_fig_p017_11.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

53 extracted references · 9 canonical work pages

  1. [1]

    , " * write output.state after.block = add.period write newline

    ENTRY address archivePrefix author booktitle chapter doi edition editor eprint howpublished institution journal key month number organization pages publisher school series title misctitle type volume year version url label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION init.state.consts ...

  2. [2]

    write newline

    " write newline "" before.all 'output.state := FUNCTION format.url url empty "" new.block "" url * "" * if FUNCTION format.eprint eprint empty "" archivePrefix empty "" archivePrefix "arXiv" = new.block " " eprint * " " * new.block " " eprint * " " * if if if FUNCTION format.doi doi empty "" " " doi * " " * if FUNCTION format.pid doi empty eprint empty ur...

  3. [3]

    HD d d d d d d d d d d d d d d uJN1 2 H d d d d d d d d d d d d d d d k ?, Ʋ ۋk

    thebibliography [1] 20pt to REFERENCES 6pt =0pt -12pt 10pt plus 3pt =0pt =0pt =1pt plus 1pt =0pt =0pt -12pt =13pt plus 1pt =20pt =13pt plus 1pt \@M =10000 =-1.0em =0pt =0pt 0pt =0pt =1.0em @enumiv\@empty 10000 10000 `\.\@m \@noitemerr \@latex@warning Empty `thebibliography' environment \@ifnextchar \@reference \@latexerr Missing key on reference command E...

    2021

  4. [4]

    N., Adelman-McCarthy , J

    Abazajian , K. N., Adelman-McCarthy , J. K., Ag \"u eros , M. A., et al. 2009, , 182, 543, 10.1088/0067-0049/182/2/543

    doi:10.1088/0067-0049/182/2/543 2009

  5. [5]

    Abbott , T. M. C., Abdalla , F. B., Allam , S., et al. 2018, , 239, 18, 10.3847/1538-4365/aae9f0

    doi:10.3847/1538-4365/aae9f0 2018

  6. [6]

    K., Glazebrook , K., Brinkmann , J., et al

    Baldry , I. K., Glazebrook , K., Brinkmann , J., et al. 2004, , 600, 681, 10.1086/380092

    doi:10.1086/380092 2004

  7. [7]

    A., Phillips , M

    Baldwin , J. A., Phillips , M. M., & Terlevich , R. 1981, , 93, 5, 10.1086/130766

    doi:10.1086/130766 1981

  8. [8]

    L., Morris , S

    Balogh , M. L., Morris , S. L., Yee , H. K. C., Carlberg , R. G., & Ellingson , E. 1999, , 527, 54, 10.1086/308056

    doi:10.1086/308056 1999

Show all 53 references
  1. [9]

    F., McIntosh , D

    Bell , E. F., McIntosh , D. H., Katz , N., & Weinberg , M. D. 2003, , 149, 289, 10.1086/378847

    2003 doi

  2. [10]

    R., & Moustakas , J

    Blanton , M. R., & Moustakas , J. 2009, , 47, 159, 10.1146/annurev-astro-082708-101734

    2009 doi

  3. [12]

    2024, , 683, A42, 10.1051/0004-6361/202348544

    Cao , J., Xu , T., Deng , Y., et al. 2024, , 683, A42, 10.1051/0004-6361/202348544

    2024 doi

  4. [13]

    J., Arag \'o n-Salamanca , A., et al

    Cheng , T.-Y., Conselice , C. J., Arag \'o n-Salamanca , A., et al. 2020, , 493, 4209, 10.1093/mnras/staa501

    2020 doi

  5. [14]

    Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. 2019, in Proceedings of the 2019 Conference of the North A merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers), ed. J. Burstein, C. Doran, & T. Solori...

    2019 doi

  6. [15]

    W., & Dambre , J

    Dieleman , S., Willett , K. W., & Dambre , J. 2015, , 450, 1441, 10.1093/mnras/stv632

    2015 doi

  7. [16]

    Dom \' nguez S \'a nchez , H., Huertas-Company , M., Bernardi , M., Tuccillo , D., & Fischer , J. L. 2018, , 476, 3661, 10.1093/mnras/sty338

    2018 doi

  8. [17]

    2021, in International Conference on Learning Representations

    Dosovitskiy, A., Beyer, L., Kolesnikov, A., et al. 2021, in International Conference on Learning Representations. https://openreview.net/forum?id=YicbFdNTTy

    2021

  9. [18]

    M., Iglesias-P \'a ramo , J., et al

    Duarte Puertas , S., Vilchez , J. M., Iglesias-P \'a ramo , J., et al. 2017, , 599, A71, 10.1051/0004-6361/201629044

    2017 doi

  10. [19]

    2024, arXiv e-prints, arXiv:2405.13491, 10.48550/arXiv.2405.13491

    Euclid Collaboration , Mellier , Y., Abdurro'uf , et al. 2024, arXiv e-prints, arXiv:2405.13491, 10.48550/arXiv.2405.13491

    2024 doi

  11. [20]

    2021, arXiv preprint arXiv:2107.01682

    Gao, X., Qian, Y., & Gao, A. 2021, arXiv preprint arXiv:2107.01682

    2021 arXiv

  12. [21]

    2022, in 2022 44th Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC), IEEE, 480--483

    Gheflati, B., & Rivaz, H. 2022, in 2022 44th Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC), IEEE, 480--483

    2022

  13. [22]

    K., & Desai , S

    Gupta , R., Srijith , P. K., & Desai , S. 2022, Astronomy and Computing, 38, 100543, 10.1016/j.ascom.2021.100543

    2022

  14. [23]

    Heckman , T. M. 1980, , 87, 152

    1980

  15. [24]

    M., Tyson , J

    Ivezi \'c , Z ., Kahn , S. M., Tyson , J. A., et al. 2019, , 873, 111, 10.3847/1538-4357/ab042c

    2019 doi

  16. [25]

    2022, in Mach

    Kang, S., Shin, M.-S., & Kim, T. 2022, in Mach. Learn. Phys. Sci. workshop

    2022

  17. [26]

    M., White , S

    Kauffmann , G., Heckman , T. M., White , S. D. M., et al. 2003 a , , 341, 33, 10.1046/j.1365-8711.2003.06291.x

    2003

  18. [27]

    M., Tremonti , C., et al

    Kauffmann , G., Heckman , T. M., Tremonti , C., et al. 2003 b , , 346, 1055, 10.1111/j.1365-2966.2003.07154.x

    2003

  19. [28]

    Kaviraj , S., Schawinski , K., Devriendt , J. E. G., et al. 2007, , 173, 619, 10.1086/516633

    2007 doi

  20. [29]

    J., Dopita , M

    Kewley , L. J., Dopita , M. A., Sutherland , R. S., Heisler , C. A., & Trevena , J. 2001, , 556, 121, 10.1086/321545

    2001 doi

  21. [31]

    2021, , 656, A140, 10.1051/0004-6361/202140757

    Kumari , N., Maiolino , R., Trussler , J., et al. 2021, , 656, A140, 10.1051/0004-6361/202140757

    2021 doi

  22. [32]

    2021, , 162, 297, 10.3847/1538-3881/ac2e96

    Lee , J., & Shin , M.-S. 2021, , 162, 297, 10.3847/1538-3881/ac2e96

    2021 doi

  23. [33]

    Y.-Y., Liao, S.-M., Huang, H.-J., Kuo, W.-T., & Ou, O

    Lin, J. Y.-Y., Liao, S.-M., Huang, H.-J., Kuo, W.-T., & Ou, O. H.-M. 2021, arXiv preprint arXiv:2110.01024

    2021 arXiv

  24. [34]

    2011, , 410, 166, 10.1111/j.1365-2966.2010.17432.x

    Lintott , C., Schawinski , K., Bamford , S., et al. 2011, , 410, 166, 10.1111/j.1365-2966.2010.17432.x

    2011

  25. [35]

    2018, Mixed Precision Training

    Micikevicius, P., Narang, S., Alben, J., et al. 2018, Mixed Precision Training. 1710.03740

    2018 arXiv

  26. [36]

    Oh , K., Sarzi , M., Schawinski , K., & Yi , S. K. 2011, , 195, 13, 10.1088/0067-0049/195/2/13

    2011 doi

  27. [37]

    K., Schawinski , K., et al

    Oh , K., Yi , S. K., Schawinski , K., et al. 2015, , 219, 1, 10.1088/0067-0049/219/1/1

    2015 doi

  28. [38]

    E., & Ferland , G

    Osterbrock , D. E., & Ferland , G. J. 2006, Astrophysics of gaseous nebulae and active galactic nuclei

    2006

  29. [39]

    2024, , 531, 4990, 10.1093/mnras/stae1450

    Parker , L., Lanusse , F., Golkar , S., et al. 2024, , 531, 4990, 10.1093/mnras/stae1450

    2024 doi

  30. [40]

    F., Jim \'e nez-Vicente , J., et al

    S \'a nchez-Bl \'a zquez , P., Peletier , R. F., Jim \'e nez-Vicente , J., et al. 2006, , 371, 703, 10.1111/j.1365-2966.2006.10699.x

    2006

  31. [41]

    2007, , 382, 1415, 10.1111/j.1365-2966.2007.12487.x

    Schawinski , K., Thomas , D., Sarzi , M., et al. 2007, , 382, 1415, 10.1111/j.1365-2966.2007.12487.x

    2007

  32. [42]

    D., & Zhang , C

    Schawinski , K., Turp , M. D., & Zhang , C. 2018, , 616, L16, 10.1051/0004-6361/201833800

    2018 doi

  33. [43]

    Schawinski , K., Zhang , C., Zhang , H., Fowler , L., & Santhanam , G. K. 2017, , 467, L110, 10.1093/mnrasl/slx008

    2017 doi

  34. [44]

    2023, , 135, 084101, 10.1088/1538-3873/ace851

    Seo , E., Kim , S., Lee , Y., et al. 2023, , 135, 084101, 10.1088/1538-3873/ace851

    2023 doi

  35. [45]

    2018, , 477, 2513, 10.1093/mnras/sty764

    Stark , D., Launet , B., Schawinski , K., et al. 2018, , 477, 2513, 10.1093/mnras/sty764

    2018 doi

  36. [46]

    R., et al

    Strateva , I., Ivezi \'c , Z ., Knapp , G. R., et al. 2001, , 122, 1861, 10.1086/323301

    2001 doi

  37. [47]

    A., Weinberg , D

    Strauss , M. A., Weinberg , D. H., Lupton , R. H., et al. 2002, , 124, 1810, 10.1086/342343

    2002 doi

  38. [48]

    2022, Injury, 53, 2625

    Tanzi, L., Audisio, A., Cirrincione, G., Aprato, A., & Vezzetti, E. 2022, Injury, 53, 2625

    2022

  39. [49]

    Tarsitano , F., Bruderer , C., Schawinski , K., & Hartley , W. G. 2022, , 511, 3330, 10.1093/mnras/stac233

    2022 doi

  40. [50]

    C., Faber , S

    Trager , S. C., Faber , S. M., Worthey , G., & Gonz \'a lez , J. J. 2000, , 119, 1645, 10.1086/301299

    2000 doi

  41. [51]

    2017, in Advances in Neural Information Processing Systems, 5998--6008

    Vaswani, A., Shazeer, N., Parmar, N., et al. 2017, in Advances in Neural Information Processing Systems, 5998--6008

    2017

  42. [52]

    Veilleux , S., & Osterbrock , D. E. 1987, , 63, 295, 10.1086/191166

    1987 doi

  43. [53]

    2020, , 491, 1554, 10.1093/mnras/stz2816

    Walmsley , M., Smith , L., Lintott , C., et al. 2020, , 491, 1554, 10.1093/mnras/stz2816

    2020 doi

  44. [54]

    M., Gonzalez , J

    Worthey , G., Faber , S. M., Gonzalez , J. J., & Burstein , D. 1994, , 94, 687, 10.1086/192087

    1994 doi

  45. [55]

    G., Adelman , J., Anderson , Jr., J

    York , D. G., Adelman , J., Anderson , Jr., J. E., et al. 2000, , 120, 1579, 10.1086/301513

    2000 doi

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.