← Back to the catalog
Private

ECOSTRESS rock vswir axis 61f98690

ecostress · other

ECOSTRESS rock vswir axis 61f98690. v2.0 standardized NIRS package: 1 spectral source(s), 3 declared target(s). Auto-generated from dataset_card.json (verify before publication).

nirv2ecostress
🔒
Private dataset. Full metadata and metrics are shown, but the bytes are not redistributed here — exporting the data requires a Dataverse token. The identity card carries no spectra, only descriptive statistics.
83
samples
2,101
wavelengths
1
sources
3
targets
27
metadata
other
family

Dataset property explorer

Mean profile risk0.42
Highest axisDistance à la référence · 1.00
Diagnostics8
Sources profiled1
ECOSTRESS rock vswir axis 61f98690 property profile0.250.50.751integritynoiseartefactsbaselinePCA outliersreferencerepeatabilitystructureECOSTRESS rock vswir axis 61f98690 profileintegrity: 0.00noise: 0.03artefacts: 0.22baseline: 0.26PCA outliers: 0.84reference: 1.00repeatability: 0.00structure: 1.00ECOSTRESS rock …0 center · 1 outer ring · outward = stronger anomaly / heterogeneity signal

Profile axes

Intégrité0.00
Artefacts locaux0.22
Bruit0.03
Outliers PCA0.84
Distance à la référence1.00
Répétabilité0.00
Baseline / forme0.26
Structure multi-régimes1.00
Diagnostic hypotheses00.250.50.751hypothesis scoreSpectre hors domaine valideSpectre hors domaine valide: 0.780.78Dataset multi-régimesDataset multi-régimes: 0.650.65Splice / raccord détecteursSplice / raccord détecteurs: 0.590.59Fond différentFond différent: 0.570.57Signature VERA25-likeSignature VERA25-like: 0.540.54Mélange feuille + fondMélange feuille + fond: 0.540.54Erreur interpolation / réécha…Erreur interpolation / rééchantillonnage: 0.500.50Erreur calibration / référenc…Erreur calibration / référence blanche: 0.500.50
DiagnosticScoreForceSignauxInterprétation probable
Spectre hors domaine valideX0.78forteRMS/SAM référence 1.00, Structure PCA 1.00, artefacts faibles 0.78Variété, espèce, lot ou condition différente mais physiquement plausible.
Dataset multi-régimesX0.65moyenneStructure PCA 1.00, RMS/SAM référence 1.00, PCA Q 0.84Mélange de campagnes, opérateurs, lots, setups ou sous-populations spectrales.
Splice / raccord détecteursX0.59moyenneRMS/SAM référence 1.00, SNR non dégradé 1.00, PCA Q 0.84Rupture aux jonctions de détecteurs, calibration locale ou sonde différente.
Fond différentX0.57moyenneRMS/SAM référence 1.00, PCA Q 0.84, Mahalanobis / T2 0.70Effet systématique du support, blanc/noir, transflectance ou environnement de mesure.
Signature VERA25-likeX0.54moyenneRMS/SAM référence 1.00, PCA Q 0.84, Mahalanobis / T2 0.70Combinaison possible changement de sonde + splice, amplifiée par géométrie, fond ou calibration.
Mélange feuille + fondX0.54moyenneRMS/SAM référence 1.00, PCA Q 0.84, Mahalanobis / T2 0.70Couverture partielle du spot; contribution du fond ou du support.
Erreur interpolation / rééchantillonnageX0.50moyenneSNR normal/élevé 1.00, Noise RMS faible 0.97, PCA Q 0.84Artefacts numériques ou traitement spectral incorrect.
Erreur calibration / référence blancheX0.50moyenneRMS/SAM référence 1.00, PCA Q 0.84, Mahalanobis / T2 0.70Décalage systématique entre campagnes, instruments ou référence blanche.

Spectral sources

rock vswir

X · other · source instruments vary by sample
rock vswir spectra0204060800123q05-q95 envelopeq25-q75 envelopemedian spectrummedianq25–q75q05–q95wavelength / none2.5none — median 15.53 (q25–q75 11.01–24.62)2.485none — median 15.95 (q25–q75 11.15–25.2)2.47none — median 16.52 (q25–q75 11.22–26.4)2.455none — median 16.99 (q25–q75 11.51–27)2.44none — median 17.69 (q25–q75 11.46–28.07)2.424none — median 18.05 (q25–q75 11.44–28.78)2.409none — median 18.69 (q25–q75 11.66–29.72)2.394none — median 18.8 (q25–q75 11.74–30.17)2.379none — median 18.68 (q25–q75 11.82–30.25)2.364none — median 18.57 (q25–q75 11.85–30.41)2.349none — median 17.8 (q25–q75 11.76–30.22)2.334none — median 17.86 (q25–q75 11.48–30.22)2.319none — median 17.93 (q25–q75 11.71–30.74)2.304none — median 18.25 (q25–q75 12.03–31.48)2.288none — median 18.89 (q25–q75 12.54–31.39)2.273none — median 20.01 (q25–q75 12.64–32.07)2.258none — median 20.04 (q25–q75 12.68–33)2.243none — median 20.34 (q25–q75 13.09–33.7)2.228none — median 20.79 (q25–q75 13.15–33.76)2.213none — median 20.43 (q25–q75 13.13–32.87)2.198none — median 20.77 (q25–q75 12.51–33.34)2.183none — median 21.39 (q25–q75 13.16–34.03)2.168none — median 21.9 (q25–q75 12.95–34.45)2.153none — median 22.29 (q25–q75 13.18–34.8)2.137none — median 22.45 (q25–q75 13.69–35.99)2.122none — median 22.38 (q25–q75 13.82–35.9)2.107none — median 22.45 (q25–q75 13.8–35.59)2.092none — median 22.55 (q25–q75 13.78–35.81)2.077none — median 22.7 (q25–q75 13.84–36.08)2.062none — median 22.74 (q25–q75 13.93–36.15)2.047none — median 22.59 (q25–q75 13.98–36.1)2.032none — median 22.41 (q25–q75 14.02–35.7)2.017none — median 22.26 (q25–q75 13.97–35.11)2.001none — median 22.16 (q25–q75 13.93–34.54)1.986none — median 22.06 (q25–q75 13.92–34.17)1.971none — median 21.8 (q25–q75 13.71–33.84)1.956none — median 21.76 (q25–q75 13.64–33.45)1.941none — median 21.66 (q25–q75 13.46–33.57)1.926none — median 21.57 (q25–q75 13.39–33.36)1.911none — median 21.5 (q25–q75 13.39–33.49)1.896none — median 21.96 (q25–q75 13.86–34.94)1.881none — median 23.29 (q25–q75 14.19–37.06)1.865none — median 23.75 (q25–q75 14.24–38.55)1.85none — median 23.95 (q25–q75 14.24–38.82)1.835none — median 24.01 (q25–q75 14.18–38.98)1.82none — median 24.21 (q25–q75 14.13–38.98)1.805none — median 24.35 (q25–q75 14.15–39.1)1.79none — median 24.3 (q25–q75 14.14–39.05)1.775none — median 24.28 (q25–q75 14.17–38.6)1.76none — median 24.19 (q25–q75 14.17–38.33)1.745none — median 24.38 (q25–q75 14.25–38.1)1.729none — median 24.2 (q25–q75 14.25–38.92)1.714none — median 24.13 (q25–q75 14.22–38.95)1.699none — median 24.31 (q25–q75 14.2–38.96)1.684none — median 24.03 (q25–q75 14.01–38.96)1.669none — median 23.94 (q25–q75 14.01–38.98)1.654none — median 23.76 (q25–q75 14–38.84)1.639none — median 23.63 (q25–q75 14.08–38.81)1.624none — median 23.46 (q25–q75 13.99–38.73)1.609none — median 23.3 (q25–q75 14.07–38.68)1.594none — median 23.13 (q25–q75 13.97–38.47)1.578none — median 23.09 (q25–q75 14–38.27)1.563none — median 23.01 (q25–q75 14.07–38.13)1.548none — median 22.9 (q25–q75 14.02–37.9)1.533none — median 22.84 (q25–q75 14.06–37.75)1.518none — median 22.75 (q25–q75 14.03–37.54)1.503none — median 22.82 (q25–q75 13.97–37.3)1.488none — median 22.78 (q25–q75 13.84–36.76)1.473none — median 22.74 (q25–q75 13.69–36.73)1.458none — median 22.62 (q25–q75 13.51–36.45)1.442none — median 22.52 (q25–q75 13.36–35.87)1.427none — median 21.92 (q25–q75 13.41–36.12)1.412none — median 21.17 (q25–q75 12.63–35.16)1.397none — median 21.69 (q25–q75 13.49–36.15)1.382none — median 22.22 (q25–q75 13.59–36.25)1.367none — median 22.1 (q25–q75 14.01–36.37)1.352none — median 22.07 (q25–q75 14.33–36.32)1.337none — median 21.92 (q25–q75 14.27–36.12)1.322none — median 21.99 (q25–q75 14.34–36.02)1.306none — median 21.77 (q25–q75 14.28–35.92)1.291none — median 21.56 (q25–q75 14.2–35.87)1.276none — median 21.53 (q25–q75 14.24–35.9)1.261none — median 21.36 (q25–q75 14.12–35.95)1.246none — median 21.24 (q25–q75 14.1–35.88)1.231none — median 21.07 (q25–q75 14.08–35.79)1.216none — median 20.95 (q25–q75 14.02–35.73)1.201none — median 20.8 (q25–q75 13.98–35.62)1.186none — median 20.68 (q25–q75 13.97–35.58)1.171none — median 20.61 (q25–q75 13.92–35.66)1.155none — median 20.66 (q25–q75 13.9–35.81)1.14none — median 20.55 (q25–q75 13.86–35.95)1.125none — median 20.44 (q25–q75 13.86–36.12)1.11none — median 20.52 (q25–q75 13.62–36.27)1.095none — median 20.53 (q25–q75 13.55–36.39)1.08none — median 20.69 (q25–q75 13.01–36.56)1.065none — median 20.64 (q25–q75 12.81–36.56)1.05none — median 20.76 (q25–q75 12.8–36.33)1.035none — median 20.74 (q25–q75 12.88–36.2)1.019none — median 20.91 (q25–q75 12.79–35.98)1.004none — median 21.03 (q25–q75 12.79–35.82)0.989none — median 21.03 (q25–q75 12.66–35.32)0.974none — median 21.26 (q25–q75 12.6–34.64)0.959none — median 21.13 (q25–q75 12.52–34.78)0.944none — median 21.32 (q25–q75 12.48–34.89)0.929none — median 21.43 (q25–q75 12.41–35.02)0.914none — median 21.44 (q25–q75 12.39–35.05)0.899none — median 21.32 (q25–q75 12.3–34.96)0.883none — median 21.29 (q25–q75 12.3–34.96)0.868none — median 21.57 (q25–q75 12.32–34.95)0.853none — median 21.47 (q25–q75 12.74–35.15)0.838none — median 21.56 (q25–q75 12.81–35.15)0.823none — median 21.72 (q25–q75 12.75–35.17)0.808none — median 21.89 (q25–q75 12.69–35.13)0.793none — median 21.93 (q25–q75 12.61–34.8)0.778none — median 21.8 (q25–q75 12.57–34.43)0.763none — median 21.47 (q25–q75 12.49–33.99)0.747none — median 21.89 (q25–q75 12.48–33.56)0.732none — median 21.39 (q25–q75 12.54–33.23)0.717none — median 20.65 (q25–q75 12.49–32.9)0.702none — median 20.48 (q25–q75 12.49–32.64)0.687none — median 20.35 (q25–q75 12.36–32.4)0.672none — median 20.56 (q25–q75 12.2–32.4)0.657none — median 20.84 (q25–q75 12.22–32.19)0.642none — median 21.16 (q25–q75 12.22–31.8)0.627none — median 21.53 (q25–q75 12.2–31.38)0.612none — median 21.6 (q25–q75 12.3–30.53)0.596none — median 21.47 (q25–q75 12.32–29.18)0.581none — median 20.37 (q25–q75 12.05–28.87)0.566none — median 19.39 (q25–q75 11.56–28.47)0.551none — median 18.73 (q25–q75 11.39–27.46)0.536none — median 17.87 (q25–q75 11.12–26.74)0.521none — median 17.59 (q25–q75 10.6–25.91)0.506none — median 16.64 (q25–q75 10.33–25.3)0.491none — median 16.36 (q25–q75 10.08–24.65)0.476none — median 15.97 (q25–q75 9.929–23.91)0.46none — median 15.53 (q25–q75 9.897–23.53)0.445none — median 15.26 (q25–q75 9.914–23.19)0.43none — median 14.94 (q25–q75 9.907–22.6)0.415none — median 14.69 (q25–q75 9.772–22.09)0.4none — median 14.5 (q25–q75 9.693–21.7)

Sampling

Wavelengths2,101
Axis range0.4–2.5 none
Mean spacing0.001 none
Griduniform
Observations83

Signal & quality

Value range2.93 – 82.3
Mean range16.9 – 28.3
Mean level25.29
Area53.11
PTP11.43
Noise RMS0.0345
SNR7.3e+02
SNR dB6e+01 dB
Dynamic range11.4
Smoothness0.1322
Saturated0.0%
X-outliers38

Integrity & artefacts

NaN ratio0.00%
Inf count0
Zero ratio0.00%
Spike count385
Spike rate0.22%
Jump count347
Jump rate0.20%
Clip fraction0.00%

Shape & reference

Baseline slope1.5365
Curvature RMS0.13247
D1 RMS0.080595
RMS to mean11.577
RMS p9535.132
SAM to mean0.11934
SAM p950.27603
Affine offset p9531.745
Affine gain p95 Δ1.9997
Affine residual p956.7324
Xcorr lag p9550

Outliers & repeatability

PCA Q p95/median6.7
Hotelling T2 p95/median5.6
Mahalanobis H p95/median2.4
Repeat groups0

Dimensionality (PCA)

Effective rank1.3
PCs → 95% var2
PCs → 99% var4
Top-10 cum. var99.9%
Computed metric scores 29worst 1.00
FamilleMétrique calculéeValeurScoreNiveauInterprétation datasetCauses typiquesCalcul / scoring
Intégrité des donnéesNaN ratiointegrity.nan_ratio0%0.00faibleSpectre completErreur acquisition/exportcount(isnan(X)) / X.sizealert = min(1, nan_ratio / 0.05)
Intégrité des donnéesInf countintegrity.inf_count00.00faibleNormalCalculs invalidescount(isinf(X))alert = min(1, inf_count / 1)
Intégrité des donnéesZero ratiointegrity.zero_ratio0%0.00faibleNormalExport, saturationcount(X == 0) / count(finite X)alert = min(1, zero_ratio / 0.05)
Amplitude globaleMean reflectanceamplitude.mean_reflectance25.2890.26faibleTrop sombreFond, géométriemean(X finite)alert reuses baseline/shape drift because absolute reflectance ranges are technology-dependent
Amplitude globaleArea under curveamplitude.area_under_curve53.1150.26faibleNormalDistance sondetrapezoid(mean_spectrum, spectral_axis)alert reuses baseline/shape drift because area scale depends on axis and units
Amplitude globalePeak-to-peak (PTP)amplitude.peak_to_peak11.4340.00faibleVariabilité forteSaturationmax(mean_spectrum) - min(mean_spectrum)alert increases when dynamic range is abnormally flat
Amplitude globaleVarianceamplitude.variance272.470.00faibleNormal ou hétérogèneMauvais contactvar(X finite)alert increases when variance/dynamic range is abnormally flat
BruitNoise RMSnoise.noise_rms0.03450.03faibleStableLampe, détecteurmedian MAD(second derivative) * 1.4826 / sqrt(6)alert = noise_rms / signal_scale, saturated at 5%
BruitSNRnoise.snr733.030.00faibleBon signalAcquisitionmean(abs(X)) / noise_rmsalert decreases with SNR dB; >=40 dB is treated as low alert
BruitBandwise SNRnoise.bandwise_snr_min75.910.00faibleZone fiableDétecteurmin(abs(mean_spectrum) / local second-derivative noise)alert decreases with worst-band SNR dB; >=35 dB is treated as low alert
Artefacts locauxSpike countartefacts.spike_count3850.22faibleSpectre propreCosmic rays, splicecount robust outliers in second derivativealert follows spike_rate, saturated at 1%
Artefacts locauxSpike rateartefacts.spike_rate0.221%0.22faibleNormalInterpolationspike_count / (n_samples * (n_features - 2))alert = min(1, spike_rate / 0.01)
Artefacts locauxJump countartefacts.jump_count3470.20faibleContinuSplicecount robust outliers in first derivativealert follows jump_rate, saturated at 1%
Artefacts locauxJump rateartefacts.jump_rate0.199%0.20faibleNormalCalibrationjump_count / (n_samples * (n_features - 1))alert = min(1, jump_rate / 0.01)
Artefacts locauxClip fractionartefacts.clip_fraction0.00115%0.00faibleNormalDétecteur saturéfraction of finite cells equal to repeated min/max extremaalert = min(1, clip_fraction / 0.01)
Forme spectraleBaseline slopeshape.baseline_slope1.53650.12faibleStableÉclairementlinear slope of mean_spectrum over normalized axisalert = abs(slope / signal_scale), saturated at 0.5
Forme spectraleCurvature RMSshape.curvature_rms0.132470.52moyenForme inhabituelleFond, splicemedian RMS(second derivative per spectrum)alert = curvature_rms / signal_scale, saturated at 1%
Forme spectraleD1 RMSshape.d1_rms0.0805950.06faiblePlatBiologie ou artefactmedian RMS(first derivative per spectrum)alert = d1_rms / signal_scale, saturated at 5%
Outliers multivariésPCA Q (SPE)outliers.pca_q_ratio6.7110.84fortSpectre atypiqueArtefact, mélangep95(Q/SPE residual) / median(Q/SPE residual)alert = min(1, pca_q_ratio / 8)
Outliers multivariésHotelling T²outliers.hotelling_t2_ratio5.61780.70moyenExtrême mais cohérentVariabilité naturellep95(Hotelling T2) / median(Hotelling T2)alert = min(1, hotelling_t2_ratio / 8)
Outliers multivariésMahalanobis Houtliers.mahalanobis_h_ratio2.370.59moyenOutlier globalDomaine différentp95(sqrt(T2)) / median(sqrt(T2))alert = min(1, mahalanobis_h_ratio / 4)
Comparaison à référenceRMS to mean spectrumreference.rms_to_mean_spectrum_p9535.1321.00fortSpectre différentDomain shiftp95 RMS distance to dataset mean spectrumalert = RMS_p95 / signal_scale, saturated at 25%
Comparaison à référenceSpectral Angle Mapper (SAM)reference.sam_to_mean_spectrum_p950.276030.79fortForme différenteFond, géométriep95 spectral angle to dataset mean spectrumalert = min(1, SAM_p95 / 0.35 rad)
RépétabilitéRMS intra-IDrepeatability.rms_intra_id—0.00faibleStablePositionnementmedian RMS distance to repeated-sample centroidalert = RMS_intra_ID / signal_scale, saturated at 10%
RépétabilitéSAM intra-IDrepeatability.sam_intra_id—0.00faibleStableAcquisitionmedian SAM to repeated-sample centroidalert = min(1, SAM_intra_ID / 0.15 rad)
RépétabilitéCV intra-IDrepeatability.cv_intra_id—0.00faibleStableOpérateurmedian within-ID band CValert = min(1, CV_intra_ID / 0.25)
Structure du datasetPCA score densitystructure.pca_score_density0.00935141.00fortSous-populationsLots différents1 / median kNN distance in PCA score spacealert follows density_cv/profile structure complexity, not raw density alone
Structure du datasetLocal Outlier Factor (LOF)structure.local_outlier_factor_p954.42521.00fortSpectre isoléCas raresp95 approximate LOF from PCA-score kNN distancesalert = min(1, max(0, LOF_p95 - 1) / 2)
Structure du datasetIsolation Forest scorestructure.isolation_forest_score_p950.604021.00fortSpectre atypiqueDiverses causesp95 IsolationForest anomaly score on PCA scoresalert follows structure complexity; raw score is implementation-dependent
X PCA score plot-3,000-2,000-1,00001,0002,000-750-500-2500250500PC1 -359.6 · PC2 -66.18PC1 239 · PC2 -80.21PC1 -139.4 · PC2 -52.26PC1 214.9 · PC2 -53.81PC1 888.4 · PC2 -43.42PC1 412.7 · PC2 -54.95PC1 1003 · PC2 -17.26PC1 -610.9 · PC2 168.2PC1 174.3 · PC2 14.36PC1 -696.1 · PC2 177.4PC1 797.9 · PC2 -64.66PC1 -470.6 · PC2 -72.69PC1 -123 · PC2 -7.139PC1 -198.3 · PC2 191.5PC1 -488.2 · PC2 221.7PC1 -1571 · PC2 91.56PC1 -43.56 · PC2 -32.12PC1 -431.3 · PC2 -78.61PC1 113.3 · PC2 -60.02PC1 133.8 · PC2 -52.15PC1 321.8 · PC2 -28.7PC1 515.2 · PC2 -95.76PC1 -35.11 · PC2 175.7PC1 180.2 · PC2 -16.42PC1 255.2 · PC2 -114.8PC1 737.3 · PC2 -25.32PC1 950.6 · PC2 -35.51PC1 686.1 · PC2 -57.6PC1 267.8 · PC2 55.87PC1 659.3 · PC2 42.15PC1 787.5 · PC2 -35.32PC1 861.3 · PC2 -18.06PC1 448.2 · PC2 -19.68PC1 832.4 · PC2 -0.2796PC1 710.2 · PC2 -82.58PC1 65.02 · PC2 199.9PC1 624.1 · PC2 -60.14PC1 747.1 · PC2 40.55PC1 -392.5 · PC2 128.7PC1 732.8 · PC2 -77.93PC1 -746.2 · PC2 -46.71PC1 -1671 · PC2 -593.6PC1 -238.6 · PC2 -251.6PC1 -970.7 · PC2 -274.6PC1 731.1 · PC2 -65.16PC1 344.1 · PC2 180.1PC1 471.4 · PC2 -65.97PC1 -623.2 · PC2 -61.7PC1 437.8 · PC2 29.69PC1 598.1 · PC2 130.7PC1 566.6 · PC2 105.1PC1 -1711 · PC2 80.34PC1 355.6 · PC2 26.5PC1 207.8 · PC2 -152PC1 817.1 · PC2 -58.55PC1 -502.8 · PC2 -0.9351PC1 -867.2 · PC2 -494.4PC1 574.1 · PC2 -85.41PC1 -465.9 · PC2 35.62PC1 -1675 · PC2 74.5PC1 -78.3 · PC2 -28.83PC1 -671.5 · PC2 175.1PC1 -976.3 · PC2 -53.09PC1 110.8 · PC2 -103.7PC1 545.1 · PC2 37.25PC1 349.5 · PC2 76.94PC1 258.2 · PC2 41.23PC1 -1585 · PC2 260PC1 -903.9 · PC2 78.7PC1 237 · PC2 255.8PC1 -1139 · PC2 110.1PC1 -718.1 · PC2 138.5PC1 508.3 · PC2 -10.95PC1 383.7 · PC2 -19.94PC1 -79.99 · PC2 -34.28PC1 -81.67 · PC2 26PC1 208.6 · PC2 240.9PC1 765.2 · PC2 -64.38PC1 775.9 · PC2 17.95PC1 527.4 · PC2 -54.27PC1 235.3 · PC2 116.7PC1 -2270 · PC2 -139.7PC1 -832.6 · PC2 318PC1 (94.4%)PC2 (3.5%)83 scores
PCA explained variance0%25%50%75%100%PC1: 94.4% (cumulative 94.4%)1PC2: 3.5% (cumulative 97.9%)2PC3: 0.9% (cumulative 98.7%)3PC4: 0.7% (cumulative 99.5%)4PC5: 0.2% (cumulative 99.6%)5PC6: 0.1% (cumulative 99.8%)6PC7: 0.1% (cumulative 99.8%)7PC8: 0.1% (cumulative 99.9%)8PC9: 0.0% (cumulative 99.9%)9PC10: 0.0% (cumulative 99.9%)10cumulative explained variancePC variancecumulativeprincipal component · cumulative (dashed)

Metric interpretation reference

Metric catalog 29
FamilleMétriqueCe qu’elle détecteForte valeur =Faible valeur =Causes typiquesCalcul / score
Intégrité des donnéesNaN ratioDonnées manquantesSpectre corrompuSpectre completErreur acquisition/exportcount(isnan(X)) / X.sizealert = min(1, nan_ratio / 0.05)
Intégrité des donnéesInf countValeurs infiniesCorruptionNormalCalculs invalidescount(isinf(X))alert = min(1, inf_count / 1)
Intégrité des donnéesZero ratioColonnes ou cellules nullesSpectre tronquéNormalExport, saturationcount(X == 0) / count(finite X)alert = min(1, zero_ratio / 0.05)
Amplitude globaleMean reflectanceNiveau moyenTrop clair / fond visibleTrop sombreFond, géométriemean(X finite)alert reuses baseline/shape drift because absolute reflectance ranges are technology-dependent
Amplitude globaleArea under curveIntensité globaleDifférence d'éclairementNormalDistance sondetrapezoid(mean_spectrum, spectral_axis)alert reuses baseline/shape drift because area scale depends on axis and units
Amplitude globalePeak-to-peak (PTP)DynamiqueVariabilité forteSpectre platSaturationmax(mean_spectrum) - min(mean_spectrum)alert increases when dynamic range is abnormally flat
Amplitude globaleVarianceVariabilité spectraleNormal ou hétérogèneSpectre platMauvais contactvar(X finite)alert increases when variance/dynamic range is abnormally flat
BruitNoise RMSBruit haute fréquenceBruitéStableLampe, détecteurmedian MAD(second derivative) * 1.4826 / sqrt(6)alert = noise_rms / signal_scale, saturated at 5%
BruitSNRQualité signalBon signalMauvais signalAcquisitionmean(abs(X)) / noise_rmsalert decreases with SNR dB; >=40 dB is treated as low alert
BruitBandwise SNRBruit localiséZone fiableZone problématiqueDétecteurmin(abs(mean_spectrum) / local second-derivative noise)alert decreases with worst-band SNR dB; >=35 dB is treated as low alert
Artefacts locauxSpike countPics étroitsArtefactsSpectre propreCosmic rays, splicecount robust outliers in second derivativealert follows spike_rate, saturated at 1%
Artefacts locauxSpike rateDensité de picsSpectre suspectNormalInterpolationspike_count / (n_samples * (n_features - 2))alert = min(1, spike_rate / 0.01)
Artefacts locauxJump countDiscontinuitésRaccord détecteurContinuSplicecount robust outliers in first derivativealert follows jump_rate, saturated at 1%
Artefacts locauxJump rateFréquence de sautsProblème spectralNormalCalibrationjump_count / (n_samples * (n_features - 1))alert = min(1, jump_rate / 0.01)
Artefacts locauxClip fractionSaturationClippingNormalDétecteur saturéfraction of finite cells equal to repeated min/max extremaalert = min(1, clip_fraction / 0.01)
Forme spectraleBaseline slopePente globaleDériveStableÉclairementlinear slope of mean_spectrum over normalized axisalert = abs(slope / signal_scale), saturated at 0.5
Forme spectraleCurvature RMSCourbureForme inhabituelleLisseFond, splicemedian RMS(second derivative per spectrum)alert = curvature_rms / signal_scale, saturated at 1%
Forme spectraleD1 RMSVariabilité localeSpectre structuréPlatBiologie ou artefactmedian RMS(first derivative per spectrum)alert = d1_rms / signal_scale, saturated at 5%
Outliers multivariésPCA Q (SPE)Non expliqué par PCASpectre atypiqueConformeArtefact, mélangep95(Q/SPE residual) / median(Q/SPE residual)alert = min(1, pca_q_ratio / 8)
Outliers multivariésHotelling T²Extrême dans PCAExtrême mais cohérentCentralVariabilité naturellep95(Hotelling T2) / median(Hotelling T2)alert = min(1, hotelling_t2_ratio / 8)
Outliers multivariésMahalanobis HDistance au nuageOutlier globalPopulation normaleDomaine différentp95(sqrt(T2)) / median(sqrt(T2))alert = min(1, mahalanobis_h_ratio / 4)
Comparaison à référenceRMS to mean spectrumDistance moyenneSpectre différentTypiqueDomain shiftp95 RMS distance to dataset mean spectrumalert = RMS_p95 / signal_scale, saturated at 25%
Comparaison à référenceSpectral Angle Mapper (SAM)Différence de formeForme différenteSimilaireFond, géométriep95 spectral angle to dataset mean spectrumalert = min(1, SAM_p95 / 0.35 rad)
RépétabilitéRMS intra-IDReproductibilitéMauvaise répétabilitéStablePositionnementmedian RMS distance to repeated-sample centroidalert = RMS_intra_ID / signal_scale, saturated at 10%
RépétabilitéSAM intra-IDVariation de formeInstableStableAcquisitionmedian SAM to repeated-sample centroidalert = min(1, SAM_intra_ID / 0.15 rad)
RépétabilitéCV intra-IDVariabilité interneMauvais contrôleStableOpérateurmedian within-ID band CValert = min(1, CV_intra_ID / 0.25)
Structure du datasetPCA score densityClustersSous-populationsHomogèneLots différents1 / median kNN distance in PCA score spacealert follows density_cv/profile structure complexity, not raw density alone
Structure du datasetLocal Outlier Factor (LOF)Anomalie localeSpectre isoléPopulation normaleCas raresp95 approximate LOF from PCA-score kNN distancesalert = min(1, max(0, LOF_p95 - 1) / 2)
Structure du datasetIsolation Forest scoreAnomalie globaleSpectre atypiqueNormalDiverses causesp95 IsolationForest anomaly score on PCA scoresalert follows structure complexity; raw score is implementation-dependent
Technology-specific extensions
TechnologieAdaptations / métriquesAnomalies cibléesCommentaire pratique
UV-Vis 300-1000 nmBaseline, pente globale, dérive aux bords 300-350 et 900-1000; métriques par zonesLumière parasite, mauvais blanc, saturation, faible signal aux extrémitésLes bords sont souvent instables; calculer aussi des scores edge/middle.
UV-Vis 300-1000 nmSaturation / clipping proche absorbance max ou réflectance maxSignal écrêtéTrès important si absorption forte.
UV-Vis 300-1000 nmRed-edge, position de maximum, ratios de bandes si végétalDécalage biologique ou artefact optiqueAide à distinguer changement réel et problème d'acquisition.
UV-Vis 300-1000 nmSmoothness / roughness indexBruit haute fréquenceSouvent plus informatif que le SNR seul.
MIR / ATR-FTIRATR contact quality index: intensité globale, aire totale, profondeur des bandes clésMauvais contact cristal-échantillonCrucial: beaucoup d'anomalies viennent du contact ATR.
MIR / ATR-FTIRCO2 / H2O atmospheric bandsMauvaise correction atmosphériquePics parasites fréquents.
MIR / ATR-FTIRBaseline curvature / rubber-band residualDiffusion, contact, dérive baselineTrès utile avant PCA.
MIR / ATR-FTIRPeak position shiftMauvais alignement spectral / calibrationImportant en FTIR car de petits shifts comptent.
MIR / ATR-FTIRBand area ratios sur bandes connuesSpectre chimiquement incohérentÀ adapter par matrice: polysaccharides, protéines, lipides, etc.
HS-MSTotal Ion Current (TIC), Base Peak Intensity (BPI)Injection faible, ionisation instableÉquivalent MS du niveau global spectral.
HS-MSNombre de pics détectésSpectre pauvre ou trop bruitéTrop peu = mauvais signal; trop = bruit/contamination.
HS-MSMass accuracy / m/z driftProblème calibration masseFondamental en HRMS.
HS-MSRetention time drift si LC/GC-MSDérive chromatographiqueÀ suivre sur standards/QC pools.
HS-MSBlank contamination scoreContaminants / carry-overComparer échantillons vs blancs.
HS-MSInternal standard CVVariabilité instrumentaleTrès robuste si standards disponibles.
HS-MSMissingness par featureInstabilité de détectionCrucial pour filtrer les variables.
Avec répétitionsRMS intra-échantillonRépétabilité globaleApplicable à toutes les technologies.
Avec répétitionsSAM / corrélation intra-échantillonRépétabilité de formeTrès utile pour spectres.
Avec répétitionsCV intra-échantillon par bande / featureRépétabilité localeDétecte les zones instables.
Avec répétitionsICC ou variance componentsPart variance échantillon vs techniqueTrès utile si plusieurs répétitions par sample.
Avec répétitionsDistance au centroïde intra-IDRépétition aberrantePermet de flagger la mauvaise répétition plutôt que le sample entier.
Bug-hunting / supervised audits
Famille de bug potentielMéthodes à ajouterCe que ça détecteÉtat dans l’explorateur
Shift spectral globalCorrélation spectre moyen inter-dataset, DTW, cross-correlation, comparaison positions de picsDécalage en longueur d'onde, mauvais alignement, interpolation différentePartiellement calculé: cross-correlation lag et dispersion des positions de pics vs spectre moyen.
Baseline / offset / gainRégression chaque spectre vs spectre moyen: x = a + b ref + residual; suivi de a, b, RMS résiduelOffset additif, effet multiplicatif, dérive de baselineCalculé dans reference.affine_*.
Mélange de lignes / mauvais appariement X-M-YVérification index, hash des lignes, duplication ID, distance spectrale intra-ID, labels incohérentsLignes mélangées, metadata mal alignées, Y attribué au mauvais spectrePartiellement couvert par répétabilité intra-ID; checks index/hash à ajouter au pipeline canonical.
Fuite d'information / répétitions mal splitéesGroupKFold par sample_id vs StratifiedKFold random; audit des partitions par sample_idPerformance artificiellement bonne due aux répétitionsNécessite splits et benchmark modèle; non calculé par la carte descriptive.
Label bugsÉchantillons proches en X mais Y différents, confident learning, erreurs systématiques FP/FNY inversés, erreurs de saisie, classes ambiguësNécessite Y et/ou modèle; recommandé pour l'explorateur supervisé.
Sous-domaines cachésPCA/UMAP/t-SNE + clustering non supervisé + association avec dataset/Y/date/operatorLots, campagnes, sondes, backgrounds non renseignésPartiellement calculé par structure PCA/LOF; UMAP/t-SNE hors carte statique.
Artefacts localisés inconnusCarte wavelength x dataset: différence moyenne, différence variance, KS par longueur d'ondeRégions spectrales anormales non anticipéesÀ calculer au niveau banque quand plusieurs datasets partagent un axe spectral.
Ruptures instrumentalesDiscontinuités dans dérivées, changepoint detectionSplice, raccord détecteur, saut local non prévuCalculé par jump/spike rates; changepoint plus avancé à ajouter.
Mélange / contamination spectraleNMF / unmixing / reconstruction par convex hullComposante externe: fond, plastique, solNon calculé automatiquement; nécessite hypothèses de composants ou grande bibliothèque.
Features instables mais prédictivesImportance modèle vs instabilité QC par variableModèle qui apprend un artefact plutôt qu'un signal biologiqueNécessite modèle supervisé; recommandé pour rapports de benchmark.

Variables

Targets 3

material_name

target · categorical
material_name classesQuartziteQuartzite: 22Quartz ConglomerateQuartz Conglomerate: 22TrachyteTrachyte: 11SyeniteSyenite: 11Hornblende SyeniteHornblende Syenite: 11Nepheline-Sodalite SyeniteNepheline-Sodalite Syenite: 11ScoriaScoria: 11Amygdaloidal BasaltAmygdaloidal Basalt: 11ObsidianObsidian: 11TuffTuff: 11+10 more+10 more: 1010
n / missing83 / 0
Classes81
Balance (entropy)1
Imbalance ratio2
Top classQuartzite (2)

class_label

target · categorical
class_label classesIgneousIgneous: 3838SedimentarySedimentary: 2828MetamorphicMetamorphic: 1717
n / missing83 / 0
Classes3
Balance (entropy)0.95
Imbalance ratio2
Top classIgneous (38)

subclass

target · categorical
subclass classesFelsicFelsic: 1111MaficMafic: 99LimestoneLimestone: 99IntermediateIntermediate: 77SandstoneSandstone: 77AlkalicAlkalic: 44UltramaficUltramafic: 44MarbleMarble: 44SchistSchist: 44ShaleShale: 44+10 more+10 more: 1919
n / missing83 / 0
Classes21
Balance (entropy)0.92
Imbalance ratio11
Top classFelsic (11)

Metadata 3

ecostress_resource_id

metadata · categorical
n / missing83 / 0
Classes83
Balance (entropy)1
Imbalance ratio1
Top classrock.igneous.alkalic.solid.vswir.ward17.jpl.perkin.spectrum (1)

location

metadata · categorical
location classesWard's Collection of American…Ward's Collection of American Rocks. Sample locality - Portland, Middlesex County, Connecticut.: 22Ward's Collection of American…Ward's Collection of American Rocks. Sample locality - Mineral Hill, near Cripple Creek, Teller County, Colorado.: 11Ward's Collection of American…Ward's Collection of American Rocks. Sample locality - Golden Cycle Mine, near Victor, Teller County, Colorado.: 11Ward's Collection of American…Ward's Collection of American Rocks. Sample locality - Near Cuttingsville, Ruthland County, Vermont.: 11Ward's Collection of American…Ward's Collection of American Rocks. Sample locality - Near Red Hill, Carroll County, New Hampshire.: 11Ward's Collection of American…Ward's Collection of American Rocks. Sample locality - Klamath Falls, Klamath County, Oregon.: 11Ward's Collection of American…Ward's Collection of American Rocks. Sample locality - Mohawk Mine, Ahmeek, Keweenaw County, Michigan.: 11Ward's Collection of American…Ward's Collection of American Rocks. Sample locality - Lake County, Oregon.: 11Ward's Collection of American…Ward's Collection of American Rocks. Sample locality - Cripple Creek, Teller county, Colorado.: 11Ward's Collection of American…Ward's Collection of American Rocks. Sample locality - Mattapan, Suffolk County, Massachusetts.: 11+10 more+10 more: 1010
n / missing83 / 0
Classes82
Balance (entropy)1
Imbalance ratio2
Top classWard's Collection of American Rocks. Sample locality - Portland, Middlesex County, Connecticut. (2)

sample_description

metadata · categorical
n / missing83 / 0
Classes83
Balance (entropy)1
Imbalance ratio1
Top classA very fine-grained, porphyritic rock, light gray in color.Geologic age - Tertiary (Post-Oligocene).Collected by: Ward's Natural Science Original ASTER Spectral Library name was jpl.perkin.rock.igneous.alkalic.solid.ward17.spectrum.txt (1)
Constant metadata 15
  • categoryrock
  • material_typeRock
  • instrumentjpl.perkin
  • acquisition_modeHemispherical reflectance
  • signal_typeReflectance (percent)
  • axis_unitWavelength (micrometers)
  • axis_min0.4
  • axis_max2.5
  • n_points_original2,101
  • publication_doi10.1016/j.rse.2019.05.015
  • citationMeerdink et al. 2019, Baldridge et al. 2009
  • licenseCopyright California Institute of Technology / JPL, all rights reserved
  • rights_statusmanual_review_needed
  • usage_scopeprivate_use_only
  • notesnone

9 variable(s) omitted (no recorded values).

Alignment

Alignment levelobservation
Sample id availableyes
Samples83
Observations (total)83
Reps per samplemin 1 · mean 1 · max 1

Provenance & citation

ContributorECOSTRESS Spectral Library
Origin · url [open]https://speclib.jpl.nasa.gov/download
Origin · url [open]https://speclib.jpl.nasa.gov/
Origin · script [manual]source_to_standard.py — standardization script (maintainer-only)
Publication10.1016/j.rse.2019.05.015 — The ECOSTRESS spectral library version 1.0
Publication10.1016/j.rse.2008.11.007 — The ASTER Spectral Library Version 2.0

Governance & integrity

Tierprivate
LicenseLicenseRef-not-cleared
Permitted useResearch and benchmarking; private use only.
Access policyManual download / private-use-only per source.
RedistributionOfficial ECOSTRESS page requests citation and states copyright/all rights reserved; converted matrices are private/internal until redistribution rights are clarified.
Content version1.0.0
Schema / protocol2.0
Content hash9ccde633e2fb0c16…
Processing hash7c242ccf020749ba…
Metadata hashf6641fb28a92e877…

Load this dataset

# pip install nirs4all-datasets
from nirs4all_datasets import get

# private dataset — export requires a Dataverse token
ds = get("ecostress_rock_vswir_2101points", token="…")
X, y = ds.x(), ds.y()
print(X.shape, y.shape)

Metadata downloads are available for public datasets only. The dataset bytes are never served here — fetch them from the origin / DOI above.