Pith. sign in

REVIEW 3 major objections 5 minor 1 cited by

Deriving the stellar labels of LAMOST spectra with Stellar LAbel Machine (SLAM)

T0 review · 3 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

Pith's one-line read SLAM trains a per-pixel support-vector regression on survey spectra to derive stellar labels for LAMOST DR5, reporting cross-validated scatters of 49 K in $T_{\rm eff}$ and 0.037 dex in overall metallicity, and producing a catalog of…

desk verdict SLAM is a well-engineered SVR labeler with released code and a useful K-giant catalog, but the APOGEE-transfer validation is not truly held out and needs to be rerun before the headline scatters are trusted. read the letter →

arxiv 1908.08677 v2 pith:BG5I3H3M submitted 2019-08-23 astro-ph.SR astro-ph.GAastro-ph.IM

classification astro-ph.SRastro-ph.GAastro-ph.IM
keywords stellarlabelssupportvectorregressiondata-drivenspectroscopyLAMOSTDR5APOGEEDR15Kgiantstarsabundancescatalogs
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

The paper introduces SLAM, a data-driven method that derives stellar labels from low-resolution spectra by training a support-vector regression model on spectra whose labels are already known. The central claim is that SLAM recovers the labels of LAMOST DR5 spectra with cross-validated scatters of about 49 K in $T_{\rm eff}$, 0.10 dex in $\log g$, 0.037 dex in overall metallicity, and 0.026, 0.058, and 0.106 dex in $[\alpha/{\rm M}]$, $[{\rm C/M}]$, and $[{\rm N/M}]$ when trained on APOGEE DR15 labels. Because the model can follow highly non-linear flux changes, it handles a wider range of spectral types than quadratic data-driven models such as The Cannon. A side product is a catalog of roughly one million LAMOST K giant stars labeled with all six parameters. If the claimed scatter holds, SLAM offers a practical route for transferring precise labels from a high-resolution survey onto a much larger low-resolution sample.

What carries the argument

The central object is a support-vector regression with a radial basis function kernel, applied independently to each wavelength pixel of normalized spectra standardized to zero mean and unit variance. At every pixel, the model chooses its own complexity by scanning a small grid of penalty and kernel-width hyper-parameters and keeping the set with the lowest k-fold cross-validated mean squared error; this per-pixel adaptivity lets the flux model bend sharply at line cores while staying smooth in continuum regions. Prediction maximizes a Gaussian likelihood with a Levenberg-Marquardt optimizer, using the cross-validated model error as the flux uncertainty and the nearest training spectrum for initialization. The paper also defines coefficients of dependence that decompose each pixel's explained variance among stellar labels, which reveal that Balmer lines carry most of the temperature information and the Mg triplet region carries much of the gravity information.

What would settle it

Take a sample of the catalog's K giants with independent measurements, such as surface gravities from stellar pulsation frequencies or abundances from high-resolution spectra, and compare them with the SLAM labels; if the scatter against these independent values is substantially larger than the reported cross-validated scatters of 49 K, 0.10 dex, 0.037 dex, and so on, the claimed precision was an artifact of training and validation labels sharing the same errors.

Watch

Extended reading notes

Core claim

On its own terms, the paper establishes that a per-pixel support-vector regression generative model, whose complexity is chosen by cross-validation rather than by the user, can serve as a general stellar label estimator for LAMOST-class spectra. Trained on 17,175 common stars between LAMOST DR5 spectra and APOGEE DR15 labels, SLAM converges on more than five million LAMOST DR5 spectra and, after an empirical K-giant selection, labels about one million red giants. The reported cross-validated scatters at ${\rm SNR}_g$ around 100 are 49 K in $T_{\rm eff}$, 0.10 dex in $\log g$, 0.037 dex in $[{\rm M/H}]$, 0.026 dex in $[\alpha/{\rm M}]$, 0.058 dex in $[{\rm C/M}]$, and 0.106 dex in $[{\rm N/M}]$, which the paper presents as comparable to other up-to-date data-driven models. On the LAMOST-only training set, SLAM's per-pixel fitting error is much lower than that of a quadratic flux model, and the paper argues that the cross-validated scatter, not the formal fit error, is the honest measure of precision because training-label errors set a floor on how small that scatter can be.

Load-bearing premise

The method's measured precision is only as good as the survey labels it trains on, and the paper itself notes that label errors set a floor on the cross-validated scatter; systematic errors in the APOGEE or LAMOST labels would be inherited by both the reported scatters and the one-million-star catalog.

Editorial extensions

If this is right

  • LAMOST DR5's roughly nine million spectra become tractable: SLAM converged on more than five million of them, with non-convergence concentrated in the lowest signal-to-noise cases.
  • The catalog of about one million K giants with six stellar labels provides a large sample for studies of the Galaxy's stellar populations.
  • The paper's empirical error calibration ties catalog uncertainties to ${\rm SNR}_g$, and it advises using the carbon and nitrogen abundances only for stars with ${\rm SNR}_g > 40$.
  • Because SLAM is open-source software, the same training-transfer recipe can be rerun when either survey releases updated labels.
  • The coefficient-of-dependence diagnostic identifies which spectral regions constrain each label, which can inform line selection in future pipelines.

Reading between the lines

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

  • The reported scatters are measured against the same APOGEE labels used for training, so they quantify precision relative to those labels; an independent check against surface gravities from stellar pulsations or high-resolution abundances would reveal whether the true errors are larger.
  • The common-star transfer strategy could be generalized to other survey pairs, making SLAM a generic translator of labels from high-resolution spectra onto lower-resolution surveys whenever the target stars lie inside the training parameter space.
  • A testable extension is to add photometry or a Galactic prior to the likelihood, which the paper leaves uniform; external constraints would plausibly reduce the low-SNR biases it reports.
  • The coefficient-of-dependence diagnostic could be used to prune wavelength pixels before training, potentially easing the computational cost that grows superlinearly with the number of training spectra.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 5 minor

Summary. The paper introduces SLAM, a data-driven stellar label estimator built on support vector regression with per-pixel adaptive model complexity selected by k-fold cross-validation. The method is applied to LAMOST DR5 spectra, first using LASP stellar labels over a wide Teff range (4000-8000 K) and then using APOGEE DR15 labels for LAMOST-APOGEE common stars to predict Teff, logg, [M/H], [alpha/M], [C/M], and [N/M]. The authors report cross-validated scatters at high SNRg of about 49 K, 0.10 dex, 0.037 dex, 0.026 dex, 0.058 dex, and 0.106 dex for those labels, compare SLAM to The Cannon, and release a catalog of roughly one million LAMOST DR5 K giants with predicted labels and error estimates.

Significance. If the reported scatters are genuine held-out predictive errors, SLAM is a competitive, open-source data-driven method whose wide Teff coverage is a practical advantage for low-resolution surveys. The public code, the reproducible training procedure, and the delivered K-giant catalog are concrete contributions. The CODs in Section 6 also provide a useful interpretive check that the model is learning physically sensible wavelength-label associations. The main significance caveat is that the headline accuracies are measured relative to the same pipelines that supply the training labels, and the Section 4 validation does not unambiguously exclude training stars, so the quoted numbers should be treated as pipeline-relative until that ambiguity is resolved.

major comments (3)
  1. [§4.2–4.3, Figs. 7–8] The Section 4 performance assessment does not describe a held-out test set. The text states that 17,175 training stars are selected from the LAMOST-APOGEE common sample and that SLAM is then applied to all 8,171,443 LAMOST stars, including the 86,552 common stars; the 57,703 converged common stars used for Figures 7 and 8 therefore include the training stars as a subset. Because Section 2.3.1 defines "CV scatter" as scatter against any data set with known labels, and not specifically as prediction on excluded data, the headline values at SNRg>100 (49 K, 0.10 dex, 0.037 dex, 0.026 dex, 0.058 dex, 0.106 dex) may partly reflect how well the SVR fits its own training labels rather than genuine predictive accuracy. The authors should state explicitly whether training stars were excluded from Figures 7 and 8, and if they were not, recompute the scatters using only stars that were never used in training.
  2. [§3.3, §5.2] The validation labels are not independent ground truth: the LASP and ASPCAP labels used for the scatter measurements are the same sources as the training labels. The paper itself acknowledges in Section 3.3 that errors in the validation labels set a floor on the achievable CV scatter, and in Section 5.2 that the flux model ignores uncertainties in the training labels. This means the reported "random uncertainties" are properly scatter relative to a particular set of pipeline labels, not absolute accuracy, and the K-giant catalog inherits any systematic errors in those labels. The abstract and catalog description should either quantify the contribution of label errors (for example, with repeat observations or mock-label injection tests) or explicitly describe the scatters as pipeline-relative.
  3. [§4.3, Table 1] The headline scatter values and the fitted error-curve coefficients are quoted without any uncertainty estimates. The scatter values in Figure 8 and the coefficients a, b, and c in Table 1 are central because they become the quoted precision of the catalog, yet no bootstrap or other confidence intervals are provided for any of them. Given that the catalog errors are calibrated to these values, the authors should provide uncertainties on the scatters and on the fitted coefficients, or at least state the sample sizes used in each SNRg bin.
minor comments (5)
  1. [§2.1, Eqs. (2)–(4)] The index notation in Equations (2)–(4) is inconsistent: mu_i and s_i are written with a star index but are actually per-pixel quantities, so they should be mu_j and s_j. This makes the standardization step harder to follow.
  2. [Figure 7 caption] The caption of Figure 7 says the gray curve is "The Cannon," while the figure legend and the surrounding text indicate that the gray curve is the scatter from Ho et al. (2017). Please correct this mismatch.
  3. [Abstract and §3.3] The abstract calls the high-SNR values "random uncertainties," but Section 3.3 defines them as CV scatters against LASP labels. Using the same terminology in both places would avoid implying that these are fully independent absolute errors.
  4. [Abstract and §1] The claim that the ability to handle wide ranges of spectral types gives SLAM a "unique capability" compared to other data-driven methods is stronger than what is demonstrated; the paper compares SLAM only with The Cannon in this respect and cites, rather than benchmarks against, the Payne and other methods. I recommend softening this claim.
  5. [§4.2 and Table 2] The example catalog rows include many entries with convergence=False, yet stellar labels and errors are still listed. The reader should be told explicitly whether non-converged rows should be discarded or whether their quoted values have a different status from converged rows.

Circularity Check

1 steps flagged · score 6.0 of 10

APOGEE-based 'CV scatter' may include training stars, making the headline Section 4.3 scatters training-fit residuals rather than held-out predictions.

  1. fitted input called prediction [Section 4.2 and Section 4.3, Figures 7 and 8]
    "Then we apply the tuned SLAM model to all 8,171,443 stars ( class=STAR in LAMOST catalog) in LAMOST DR5. SLAM successfully converges for 5,132,474 stars. In the LAMOST–APOGEE common samples (86,552), SLAM converged for 57,703 of them and derived their stellar labels. ... Figure 7 shows the CV scatter of the SLAM-predicted stellar labels for the LAMOST–APOGEE common stars at different signal-to-noise ratio intervals."

    The 57,703 LAMOST–APOGEE common stars used for the Figure 7/8 scatter include the 17,175-star training sample selected in Section 4.2, and the paper never states that these training stars are excluded before computing the 'CV scatter'. Because the SVR model is trained to reproduce the APOGEE labels of exactly those stars, the SLAM-minus-APOGEE residuals for training members are training-fit residuals, not predictive errors. The headline high-SNR values (49 K in Teff, 0.10 dex in logg, 0.037 dex in [M/H], etc.) are therefore partly forced by the training objective rather than measured on held-out stars.

full rationale

The Section 3 LAMOST-only validation uses separately selected random test sets in each SNR bin and is a standard held-out test, so that part of the paper is not circular. The problem is specific to the APOGEE-transfer experiment, which carries the paper's headline accuracy claim for the ~1 million K-giant catalog. Section 4.2 trains on 17,175 common stars and then applies the model to all LAMOST stars, including the common-star sample; Section 4.3 computes 'CV scatter' from those common stars with no explicit exclusion of the training subset. Since the model's objective is to match the APOGEE labels of the training stars, the reported scatter can substantially reflect in-sample agreement. The paper's own definition of CV scatter in Section 2.3.1 is only scatter against a set with known labels, not necessarily a cross-validated held-out set, which makes the ambiguity material. The catalog-error calibration in Table 1 is then fitted to this possibly training-contaminated scatter, so the downstream error estimates inherit the same issue. This is a partial circularity: the central APOGEE-based validation numbers are not demonstrably independent predictions. It does not invalidate the method itself, which is a standard data-driven regressor, and the LAMOST-only test provides independent support for the general approach.

Assumptions & free parameters 5 free parameters · 3 assumptions · 0 invented entities

SLAM is an empirical regression method; it contributes no physical constants or entities. Its accuracy rests on the quality of survey labels, the normalization of spectra, and the generalization of the trained model to stars outside the training distribution. The main tuned quantities are SVR hyperparameters, empirical error-curve coefficients, and several hand-set thresholds and selection boundaries.

free parameters (5)
  • SVR hyperparameters C and gamma = Section 3: C in {10,100}, gamma in {0.1,0.01}; Section 4: C=10^{0..2}, gamma=10^{-3..-1}; epsilon=0.05
    Selected by cross-validation to minimize CV MSE; they set model complexity and directly affect reported scatters.
  • Error curve coefficients a, b, c = Table 1, e.g. Teff: a=204.8, b=0.056, c=38.8
    Fitted to CV scatter versus SNRg and used to produce catalog uncertainties.
  • Continuum normalization settings = 1.5 sigma exclusion threshold; bin width and spline softness adjusted by experience
    Section 2.1 says these 'can be adjusted using experience'; they affect normalized spectra used by every model.
  • Bad-pixel threshold = 50 bad pixels per spectrum
    Spectra with more than 50 bad pixels are excluded in Section 3.2 and Section 4.2.
  • K-giant selection polygon = Pink polygon in Figure 5 (empirical, not numeric)
    Defines the reliable K-giant sample in Teff-logg space; completeness ~97% is internal.
assumptions (3)
  • domain assumption Training and validation labels from LASP and ASPCAP are accurate enough to serve as ground truth
    CV scatter is computed against these labels; Section 3.3 acknowledges label errors set a floor on CV scatter and Section 5.2 notes training label uncertainties are ignored.
  • domain assumption Normalized LAMOST spectra retain stellar label information
    All predictions use pseudo-continuum normalized fluxes; Section 5.1 warns inconsistent normalization can degrade or break label estimation.
  • domain assumption The trained SVR model generalizes outside the training parameter range
    SLAM is applied to all converged LAMOST stars, including those outside the training range; Section 4.2 identifies a 'stripe' of unreliable labels and requires an empirical polygon cut.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Deriving the stellar labels of LAMOST spectra with Stellar LAbel Machine (SLAM)." pith.science (2026). https://pith.science/paper/BG5I3H3M

@misc{pith2026190808677,
  author       = {Pith},
  title        = {Pith review of: Deriving the stellar labels of LAMOST spectra with Stellar LAbel Machine (SLAM)},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/BG5I3H3M}},
  note         = {Machine review of arXiv:1908.08677}
}
abstract

The LAMOST survey has provided 9 million spectra in its Data Release 5 (DR5) at R$\sim$1800. Extracting precise stellar labels is crucial for such a large sample. In this paper, we report the implementation of the Stellar LAbel Machine (SLAM), which is a data-driven method based on Support Vector Regression (SVR), a robust non-linear regression technique. Thanks to the capability to model highly non-linear problems with SVR, SLAM generally can derive stellar labels over a wide range of spectral types. This gives it a unique capability compared to other popular data-driven methods. To illustrate this capability, we test the performance of SLAM on stars ranging from Teff$\sim$4000 to $\sim$8000 K trained on LAMOST spectra and stellar labels. At g-band signal-to-noise ratio (SNRg) higher than 100, the random uncertainties of Teff, logg and [Fe/H] are 50 K, 0.09 dex, and 0.07 dex, respectively. We then set up another SLAM model trained by APOGEE and LAMOST common stars to demonstrate its capability of dealing with high dimensional problems. The spectra are from LAMOST DR5 and the stellar labels of the training set are from APOGEE DR15, including Teff, logg, [M/H],[$\alpha$/M], [C/M], and [N/M]. The cross-validated scatters at SNRg$\sim$100 are 49 K, 0.10 dex, 0.037 dex,0.026 dex, 0.058 dex, and 0.106 dex for these parameters, respectively. This performance is at the same level as other up-to-date data-driven models. As a byproduct, we also provide the latest catalog of $\sim$1 million LAMOST DR5 K giant stars with SLAM-predicted stellar labels in this work.

Figures

Figures reproduced from arXiv: 1908.08677 by the authors.

Figure 1
Figure 1. Examples of how spectral flux changes with two primary stellar labels, i.e., Teff and log g, at two fixed wavelengths. Blue dots in left and right panels are the flux values of normalized PHOENIX spectra with [Fe/H]= 0.0 at λ = 5174 ˚A(Mg b) and 6564 ˚A(Hα), respectively. Since these two pixels are around the spectral lines which are extremely important in deriving stellar labels, a model’s fitting performance for s… view at source ↗
Figure 2
Figure 2. This figure shows the comparison of MSE and MD2 of SLAM and The Cannon. The black lines always represent the 16, 50 and 84 percentile values of each pixel of the spectra of the training set. The red lines in the upper (lower) three panels show the MSE (MD2 ) from SLAM, while the gray lines show similar quantities derived from The Cannon. The black dashed lines are the mean level of the MSE/MD2 . 3. TESTS ON LAMOST D… view at source ↗
Figure 3
Figure 3. This figure shows the distributions of the predicted stellar labels at different ranges of SNRg. The 6 rows from top to bottom correspond to 6 different SNRg intervals. In each row, the first panel shows the diagram of LAMOST DR5 Teff -log g which are regarded as the true values. The second panel shows the similar Teff and log g diagram with values derived from SLAM. The third, fourth, and the last panels show the S… view at source ↗
Figures from the paper (8 more)
Figure 4
Figure 4. Figure 4: This figure shows how the errors of stellar labels change with SNRg. In all panels, the blue curves represent the SLAM errors (formal errors). The red and purple curves represent the formal errors of tests for synthetic spectra of solar-like and K giant stars, respecti…
Figure 5
Figure 5. Figure 5: The left panel shows the distribution of SLAM-predicted Teff and log g of all converged LAMOST DR5 stars. The pink solid polygon represents the selected area for K giant stars. The middle/right panel shows LAMOST/APOGEE Teff -log g diagrams for the sample located in th…
Figure 6
Figure 6. Figure 6: The left panel shows the distribution of SLAM-predicted Teff and log g of LAMOST DR5 stars with D < 2.5. The pink solid polygon is the same as [PITH_FULL_IMAGE:figures/full_fig_p012_6.png]
Figure 7
Figure 7. Figure 7: This figure shows the comparison of the CV scatters of stellar labels between SLAM(blue) and The Cannon(gray). SLAM errors and bias are also shown with green and orange lines, respectively. The inverse signal-to-noise ratio trends are also superposed with black dashed …
Figure 8
Figure 8. Figure 8: This figure shows the diagonal plots of the 6 stellar labels (effective temperature, surface gravity, metallicity, α￾element abundance, carbon abundance, and nitrogen abundance) for the LAMOST–APOGEE common stars with SNRg > 100. 1.5 1.0 0.5 0.0 0.5 [M/H]SLAM [dex] 0.2…
Figure 9
Figure 9. Figure 9: The top-left panel shows the distribution of [α/M]-[M/H] plane for the SLAM-derived labels for the LAMOST– APOGEE common stars. The top-middle panel shows the similar plot but with APOGEE parameters. The top-right panel shows the similar plot as the top-left panel for …
Figure 10
Figure 10. Figure 10: CODs from training set with LAMOST spectra and stellar labels. The pink solid and dashed lines are the 50, 16, and 84 percentiles of normalized spectral fluxes. The blue, orange, and green filled regions represent for the COD(Teff ), COD(log g), and COD([Fe/H]), respe…
Figure 11
Figure 11. Figure 11: Two examples of how the best hyper-parameters are chosen for SVR [PITH_FULL_IMAGE:figures/full_fig_p017_11.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. A Helium-shell Burning Blue Horizontal Branch Star Produced from Common Envelope Evolution

    astro-ph.SR 2026-08 conditional novelty 6.0 of 10

    Feige 64 is a 0.35 solar mass helium-shell-burning BHB star in a 0.826-day orbit with a 1.26 solar mass white dwarf, best explained as a common-envelope remnant.

Reference graph

Works this paper leans on

22 extracted references · 18 canonical work pages · cited by 1 Pith paper

  1. [1]

    D., et al., 2017, ApJS, 233, 25 Astropy Collaboration, Robitaille, T

    Albareti F. D., et al., 2017, ApJS, 233, 25 Astropy Collaboration, Robitaille, T. P., Tollerud, E. J., et al. 2013, A&A, 558, A33 Astropy Collaboration, Price-Whelan, A. M., Sip˝ ocz, B. M., et al. 2018, AJ, 156, 123 Bailer-Jones C. A. L., 1997, PASP, 109, 932 Bailer-Jones C. A. L., Irwin M., Gilmore G., von Hippel T., 1997, MNRAS, 292, 157 Bailer-Jones C...

  2. [2]

    2018, MNRAS, 478, 4513 Brahm R., Jord´ an A., Hartman J., Bakos G., 2017, MNRAS, 467, 971

    Buder, S., Asplund, M., Duong, L., et al. 2018, MNRAS, 478, 4513 Brahm R., Jord´ an A., Hartman J., Bakos G., 2017, MNRAS, 467, 971

  3. [3]

    2015, MNRAS, 447, 256 Casey A

    Bu, Y., & Pan, J. 2015, MNRAS, 447, 256 Casey A. R., Hogg D. W., Ness M., Rix H.-W., Ho A. Q., Gilmore G., 2016, arXiv, arXiv:1603.03040 Casey A. R., et al., 2017, ApJ, 840, 59

  4. [4]

    Castelli, F., & Kurucz, R. L. 2003, Modelling of Stellar

  5. [5]

    1978, Applied Mathematical Sciences

    Atmospheres, A20 Chang., C.-C., Lin., C.-J., 2011, ACM Transactions on Intelligent Systems and Technology, 2:27:1 Cui X.-Q., et al., 2012, RAA, 12, 1197 de Boor, C. 1978, Applied Mathematical Sciences

  6. [6]

    J., Liu, C., et al

    Deng, L.-C., Newberg, H. J., Liu, C., et al. 2012, Research in Astronomy and Astrophysics, 12, 735 Freeman K. C., 2012, ASPC, 458, 393 Garc´ ıa P´ erez, A. E., Allende Prieto, C., Holtzman, J. A., et al. 2016, AJ, 151, 144 Gieseke F., et al., 2017, MNRAS, 472, 3101 Gilmore G., et al., 2012, Msngr, 147, 25 Ho A. Y. Q., et al., 2017a, ApJ, 836, 5 Ho A. Y. Q...

  7. [7]

    A., Hasselquist, S., Shetrone, M., et al

    Holtzman, J. A., Hasselquist, S., Shetrone, M., et al. 2018, AJ, 156, 125 Husser T.-O., Wende-von Berg S., Dreizler S., Homeier D., Reiners A., Barman T., Hauschildt P. H., 2013, A&A, 553, A6 Jones E, Oliphant E, Peterson P, et al. SciPy: Open Source Scientific Tools for Python, 2001, http://www.scipy.org/

  8. [8]

    2009, A&A, 501, 1269 Kuntzer T., Tewes M., Courbin F., 2016, A&A, 591, A54 Lee Y

    Koleva, M., Prugniel, P., Bouchard, A., et al. 2009, A&A, 501, 1269 Kuntzer T., Tewes M., Courbin F., 2016, A&A, 591, A54 Lee Y. S., et al., 2011, AJ, 141, 90 Lee Y. S., et al., 2008, AJ, 136, 2022 Li X.-R., Pan R.-Y., Duan F.-Q., 2017, RAA, 17, 036

Show all 22 references
  1. [9]

    Li, X., Wu, Q. M. J., Luo, A., et al. 2014, ApJ, 790, 105

  2. [10]

    2014, Research in Astronomy and Astrophysics, 14, 423-432

    Liu, C.-X., Zhang, P.-A., & Lu, Y. 2014, Research in Astronomy and Astrophysics, 14, 423-432

  3. [11]

    2015, MNRAS, 452, 1394 Liu C., Bailer-Jones C

    Lu, Y., & Li, X. 2015, MNRAS, 452, 1394 Liu C., Bailer-Jones C. A. L., Sordo R., Vallenari A., Borrachero R., Luri X., Sartoretti P., 2012, MNRAS, 426, 2463

  4. [12]

    L., et al

    Liu, C., Deng, L.-C., Carlin, J. L., et al. 2014, ApJ, 790, 110

  5. [13]

    2015, ApJ, 807, 4 Liu X.-W., et al., 2014, IAUS, 298, 310 Liu X.-W., Zhao G., Hou J.-L., 2015, RAA, 15, 1089

    Liu, C., Fang, M., Wu, Y., et al. 2015, ApJ, 807, 4 Liu X.-W., et al., 2014, IAUS, 298, 310 Liu X.-W., Zhao G., Hou J.-L., 2015, RAA, 15, 1089

  6. [14]

    2019, ApJS, 241, 32 Majewski S

    Liu, Z., Cui, W., Liu, C., et al. 2019, ApJS, 241, 32 Majewski S. R., 2012, AAS, 219, 205.06

  7. [16]

    2012, arXiv e-prints, arXiv:1201.0490

    Pedregosa, F., Varoquaux, G., Gramfort, A., et al. 2012, arXiv e-prints, arXiv:1201.0490

  8. [17]

    Perez, F., & Granger, B. E. 2007, Computing in Science and Engineering, 9, 21

  9. [18]

    2007, arXiv e-prints, astro-ph/0703658

    Prugniel, P., Soubiran, C., Koleva, M., et al. 2007, arXiv e-prints, astro-ph/0703658

  10. [19]

    2016, ApJ, 826, L25

    Rix, H.-W., Ting, Y.-S., Conroy, C., et al. 2016, ApJ, 826, L25. Skrutskie M. F., et al., 2006, AJ, 131, 1163

  11. [20]

    Smola, A. J., Sch¨ olkopf, B, Statistics and Computing, 14, 199 Soubiran C., Le Campion J.-F., Cayrel de Strobel G., Caillo A., 2010, A&A, 515, A111 Steinmetz M., et al., 2006, AJ, 132, 1645 Ting Y.-S., Conroy C., Rix H.-W., 2016, ApJ, 826, 83 Ting Y.-S., Rix H.-W., Conroy C.,...

  12. [21]

    2019, ApJ, 879, 69 Tonry J

    Ting, Y.-S., Conroy, C., Rix, H.-W., et al. 2019, ApJ, 879, 69 Tonry J. L., et al., 2012, ApJ, 750, 99 Wu Y., et al., 2011, RAA, 11, 924 Wu Y., Du B., Luo A., Zhao Y., Yuan H., 2014, IAUS, 306, 340 Xiang M. S., et al., 2015, MNRAS, 448, 822 Xiang M.-S., et al., 2017, MNRAS, 46...

  13. [22]

    2019, SLAM: Stellar LAbel Machine v1.2019.1005.0, Zenodo, doi:10.5281/zenodo.3461503

    Zhang, B. 2019, SLAM: Stellar LAbel Machine v1.2019.1005.0, Zenodo, doi:10.5281/zenodo.3461503

  14. [23]

    L., et al

    Zhong, J., Li, J., Carlin, J. L., et al. 2019, arXiv e-prints, arXiv:1908.01128

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.