← Back to the catalog
Private

ECOSTRESS manmade tir axis 85fbc8f6

ecostress · other

ECOSTRESS manmade tir axis 85fbc8f6. v2.0 standardized NIRS package: 1 spectral source(s), 1 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.
16
samples
2,256
wavelengths
1
sources
1
targets
27
metadata
other
family

Dataset property explorer

Mean profile risk0.31
Highest axisOutliers PCA · 1.00
Diagnostics8
Sources profiled1
ECOSTRESS manmade tir axis 85fbc8f6 property profile0.250.50.751integritynoiseartefactsbaselinePCA outliersreferencerepeatabilitystructureECOSTRESS manmade tir axis 85fbc8f6 profileintegrity: 0.00noise: 0.01artefacts: 0.18baseline: 0.40PCA outliers: 1.00reference: 0.16repeatability: 0.00structure: 0.75ECOSTRESS manma…0 center · 1 outer ring · outward = stronger anomaly / heterogeneity signal

Profile axes

Intégrité0.00
Artefacts locaux0.18
Bruit0.01
Outliers PCA1.00
Distance à la référence0.16
Répétabilité0.00
Baseline / forme0.40
Structure multi-régimes0.75
Diagnostic hypotheses00.250.50.751hypothesis scoreErreur interpolation / réécha…Erreur interpolation / rééchantillonnage: 0.540.54Spectre hors domaine valideSpectre hors domaine valide: 0.460.46Fond différentFond différent: 0.460.46Signature VERA25-likeSignature VERA25-like: 0.450.45Splice / raccord détecteursSplice / raccord détecteurs: 0.440.44Spectre plat / signal faibleSpectre plat / signal faible: 0.420.42Dataset multi-régimesDataset multi-régimes: 0.410.41Mélange feuille + fondMélange feuille + fond: 0.400.40
DiagnosticScoreForceSignauxInterprétation probable
Erreur interpolation / rééchantillonnageX0.54moyennePCA Q 1.00, SNR normal/élevé 1.00, Noise RMS faible 0.99Artefacts numériques ou traitement spectral incorrect.
Spectre hors domaine valideX0.46moyenneartefacts faibles 0.82, Structure PCA 0.75, Mahalanobis / T2 0.41Variété, espèce, lot ou condition différente mais physiquement plausible.
Fond différentX0.46moyennePCA Q 1.00, Mahalanobis / T2 0.41, Baseline/mean/area 0.40Effet systématique du support, blanc/noir, transflectance ou environnement de mesure.
Signature VERA25-likeX0.45moyennePCA Q 1.00, Mahalanobis / T2 0.41, Spike rate 0.18Combinaison possible changement de sonde + splice, amplifiée par géométrie, fond ou calibration.
Splice / raccord détecteursX0.44moyennePCA Q 1.00, SNR non dégradé 1.00, Spike rate 0.18Rupture aux jonctions de détecteurs, calibration locale ou sonde différente.
Spectre plat / signal faibleX0.42faiblePCA Q 1.00, artefacts faibles 0.82, Variance très faible 0.34Mauvais contact, échantillon absent, mesure dégradée ou dynamique très faible.
Dataset multi-régimesX0.41faiblePCA Q 1.00, Structure PCA 0.75, Mahalanobis / T2 0.41Mélange de campagnes, opérateurs, lots, setups ou sous-populations spectrales.
Mélange feuille + fondX0.40faiblePCA Q 1.00, Mahalanobis / T2 0.41, Baseline/mean/area 0.40Couverture partielle du spot; contribution du fond ou du support.

Spectral sources

manmade tir

X · other · source instruments vary by sample
manmade tir spectra8085909505101520q05-q95 envelopeq25-q75 envelopemedian spectrummedianq25–q75q05–q95wavelength / none15.385none — median 87.41 (q25–q75 86.38–88.4)14.688none — median 87.52 (q25–q75 86.47–88.45)14.051none — median 87.1 (q25–q75 86.25–88.21)13.432none — median 87.45 (q25–q75 86.39–88.39)12.898none — median 87.13 (q25–q75 86.19–87.96)12.404none — median 87.2 (q25–q75 86.13–87.95)11.947none — median 87.62 (q25–q75 86.31–88.22)11.496none — median 86.57 (q25–q75 85.95–87.71)11.102none — median 86.84 (q25–q75 85.78–87.54)10.735none — median 86.35 (q25–q75 85.59–87.48)10.39none — median 85.68 (q25–q75 84.99–87.33)10.068none — median 85.27 (q25–q75 84.23–86.84)9.7459none — median 85.01 (q25–q75 84.06–86.61)9.4614none — median 85.18 (q25–q75 84.19–86.7)9.193none — median 85.49 (q25–q75 84.76–87.01)8.9394none — median 85.9 (q25–q75 85.13–87.67)8.6848none — median 86.11 (q25–q75 85.32–87.45)8.4581none — median 86.75 (q25–q75 85.54–87.5)8.243none — median 87.36 (q25–q75 86.02–87.97)8.0385none — median 87.72 (q25–q75 86.29–88.2)7.8439none — median 88.49 (q25–q75 87.12–89.01)7.6472none — median 88.85 (q25–q75 87.47–89.42)7.4709none — median 88.91 (q25–q75 87.57–89.57)7.3026none — median 88.18 (q25–q75 87.05–89.07)7.1416none — median 87.86 (q25–q75 87.05–89.03)6.9782none — median 87.83 (q25–q75 87.08–89.08)6.8311none — median 87.95 (q25–q75 87.26–89.3)6.6901none — median 88.35 (q25–q75 87.28–89.21)6.5548none — median 89.01 (q25–q75 87.83–89.69)6.4248none — median 89.32 (q25–q75 88.15–89.97)6.2923none — median 88.71 (q25–q75 87.5–89.34)6.1724none — median 88.59 (q25–q75 87.38–89.26)6.0571none — median 89.36 (q25–q75 88.13–89.85)5.9459none — median 89.4 (q25–q75 88.07–89.95)5.8322none — median 89.42 (q25–q75 88.19–90.14)5.7291none — median 89.34 (q25–q75 87.94–90.29)5.6296none — median 89.29 (q25–q75 87.78–90.38)5.5334none — median 89.19 (q25–q75 87.69–90.27)5.4405none — median 89.23 (q25–q75 87.74–90.46)5.3452none — median 89.21 (q25–q75 87.65–90.48)5.2584none — median 89.26 (q25–q75 87.71–90.52)5.1745none — median 89.26 (q25–q75 87.77–90.62)5.0932none — median 89.33 (q25–q75 87.78–90.64)5.0095none — median 89.29 (q25–q75 87.81–90.67)4.9332none — median 89.33 (q25–q75 87.84–90.72)4.8593none — median 89.41 (q25–q75 87.8–90.8)4.7875none — median 89.4 (q25–q75 87.86–90.77)4.7178none — median 89.4 (q25–q75 87.88–90.82)4.6459none — median 89.41 (q25–q75 87.84–90.75)4.5802none — median 89.41 (q25–q75 87.88–90.75)4.5164none — median 89.4 (q25–q75 87.87–90.77)4.4543none — median 89.36 (q25–q75 87.85–90.79)4.3902none — median 89.4 (q25–q75 87.8–90.76)4.3315none — median 89.23 (q25–q75 87.78–90.57)4.2744none — median 88.83 (q25–q75 87.73–89.73)4.2187none — median 88.9 (q25–q75 87.77–89.81)4.1645none — median 89.36 (q25–q75 87.82–90.7)4.1084none — median 89.35 (q25–q75 87.72–90.7)4.057none — median 89.32 (q25–q75 87.77–90.58)4.0068none — median 89.19 (q25–q75 87.67–90.37)3.9579none — median 89.13 (q25–q75 87.64–90.03)3.9072none — median 89.17 (q25–q75 87.64–90.31)3.8606none — median 89.11 (q25–q75 87.67–90.27)3.8152none — median 89.2 (q25–q75 87.7–90.35)3.7708none — median 89.26 (q25–q75 87.74–90.43)3.7274none — median 89.16 (q25–q75 87.62–90.44)3.6824none — median 89.2 (q25–q75 87.68–90.44)3.641none — median 89.16 (q25–q75 87.66–90.32)3.6006none — median 89.15 (q25–q75 87.61–90.24)3.561none — median 89.15 (q25–q75 87.59–90.16)3.5199none — median 89.1 (q25–q75 87.53–89.92)3.4821none — median 89.05 (q25–q75 87.58–89.59)3.4451none — median 88.95 (q25–q75 87.54–89.56)3.4088none — median 88.94 (q25–q75 87.36–89.5)3.3712none — median 88.94 (q25–q75 87.54–89.51)3.3364none — median 88.99 (q25–q75 87.53–89.77)3.3024none — median 89.01 (q25–q75 87.58–89.79)3.2691none — median 89.1 (q25–q75 87.59–89.74)3.2365none — median 88.89 (q25–q75 87.49–89.54)3.2025none — median 89 (q25–q75 87.54–89.65)3.1711none — median 88.88 (q25–q75 87.44–89.45)3.1404none — median 88.95 (q25–q75 87.49–89.55)3.1103none — median 88.86 (q25–q75 87.34–89.47)3.0789none — median 88.9 (q25–q75 87.41–89.44)3.0499none — median 88.86 (q25–q75 87.33–89.37)3.0215none — median 88.78 (q25–q75 87.22–89.28)2.9935none — median 88.81 (q25–q75 87.27–89.16)2.9661none — median 88.7 (q25–q75 87.23–89.24)2.9376none — median 88.63 (q25–q75 87.21–89.27)2.9112none — median 88.64 (q25–q75 87.17–89.29)2.8853none — median 88.64 (q25–q75 87.09–89.36)2.8598none — median 88.74 (q25–q75 87.34–89.46)2.8332none — median 88.67 (q25–q75 87.17–89.39)2.8087none — median 88.48 (q25–q75 87.16–89.35)2.7845none — median 88.17 (q25–q75 87.04–89.14)2.7608none — median 88.34 (q25–q75 87.09–89.24)2.7375none — median 88.8 (q25–q75 87.32–89.4)2.7132none — median 88.95 (q25–q75 87.38–89.6)2.6906none — median 89.04 (q25–q75 87.49–90.01)2.6685none — median 89.35 (q25–q75 87.82–90.41)2.6467none — median 89.17 (q25–q75 87.43–90.18)2.6239none — median 89.16 (q25–q75 87.47–89.95)2.6028none — median 89.19 (q25–q75 87.65–90.22)2.5821none — median 89.22 (q25–q75 87.6–90.24)2.5617none — median 89.33 (q25–q75 87.76–90.35)2.5416none — median 89.11 (q25–q75 87.4–90.06)2.5206none — median 89.05 (q25–q75 87.35–90.06)2.5011none — median 89.01 (q25–q75 87.42–90.05)2.482none — median 88.95 (q25–q75 87.41–89.95)2.4631none — median 89 (q25–q75 87.3–89.98)2.4434none — median 88.93 (q25–q75 87.21–89.96)2.4251none — median 88.97 (q25–q75 87.29–89.99)2.4071none — median 89.08 (q25–q75 87.23–89.94)2.3893none — median 88.97 (q25–q75 87.19–89.82)2.3718none — median 88.9 (q25–q75 87.25–89.97)2.3535none — median 89 (q25–q75 87.31–89.87)2.3366none — median 89.03 (q25–q75 87.11–89.67)2.3198none — median 88.98 (q25–q75 87.15–89.87)2.3034none — median 88.99 (q25–q75 87.11–89.83)2.2861none — median 88.96 (q25–q75 87.25–89.98)2.2701none — median 88.9 (q25–q75 87.04–89.81)2.2543none — median 88.92 (q25–q75 87.2–89.82)2.2387none — median 88.95 (q25–q75 87.25–89.74)2.2233none — median 88.83 (q25–q75 87.02–89.61)2.2073none — median 88.88 (q25–q75 87.15–89.86)2.1923none — median 88.84 (q25–q75 87.2–89.66)2.1776none — median 88.97 (q25–q75 87.15–89.77)2.1631none — median 88.81 (q25–q75 87.09–89.66)2.1478none — median 88.74 (q25–q75 86.94–89.6)2.1337none — median 88.77 (q25–q75 87.18–89.71)2.1197none — median 88.77 (q25–q75 86.99–89.76)2.1059none — median 88.78 (q25–q75 87–89.65)2.0924none — median 88.86 (q25–q75 87.16–89.68)2.0781none — median 88.75 (q25–q75 87.06–89.46)2.0648none — median 88.81 (q25–q75 86.89–89.43)2.0518none — median 88.83 (q25–q75 87.03–89.52)2.0389none — median 88.82 (q25–q75 87.05–89.73)2.0253none — median 88.67 (q25–q75 87.03–89.76)2.0128none — median 88.7 (q25–q75 87–89.52)2.0003none — median 88.58 (q25–q75 86.69–89.3)

Sampling

Wavelengths2,256
Axis range2–15.39 none
Mean spacing0.00594 none
Gridirregular
Observations16

Signal & quality

Value range77.7 – 93
Mean range84.9 – 89.5
Mean level88.26
Area1171
PTP4.626
Noise RMS0.034545
SNR2.6e+03
SNR dB7e+01 dB
Dynamic range4.63
Smoothness0.1212
Saturated0.0%
X-outliers5

Integrity & artefacts

NaN ratio0.00%
Inf count0
Zero ratio0.00%
Spike count65
Spike rate0.18%
Jump count62
Jump rate0.17%
Clip fraction0.01%

Shape & reference

Baseline slope-2.5324
Curvature RMS0.11888
D1 RMS0.09407
RMS to mean1.3344
RMS p953.602
SAM to mean0.005949
SAM p950.0086084
Affine offset p9559.43
Affine gain p95 Δ0.65443
Affine residual p950.52638
Xcorr lag p950

Outliers & repeatability

PCA Q p95/median8.5
Hotelling T2 p95/median2.7
Mahalanobis H p95/median1.6
Repeat groups0

Dimensionality (PCA)

Effective rank1.3
PCs → 95% var2
PCs → 99% var2
Top-10 cum. var100.0%
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_reflectance88.2590.40faibleTrop sombreFond, géométriemean(X finite)alert reuses baseline/shape drift because absolute reflectance ranges are technology-dependent
Amplitude globaleArea under curveamplitude.area_under_curve1170.60.40faibleNormalDistance 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_peak4.62590.34faibleVariabilité forteSaturationmax(mean_spectrum) - min(mean_spectrum)alert increases when dynamic range is abnormally flat
Amplitude globaleVarianceamplitude.variance5.58650.34faibleNormal ou hétérogèneMauvais contactvar(X finite)alert increases when variance/dynamic range is abnormally flat
BruitNoise RMSnoise.noise_rms0.0345450.01faibleStableLampe, détecteurmedian MAD(second derivative) * 1.4826 / sqrt(6)alert = noise_rms / signal_scale, saturated at 5%
BruitSNRnoise.snr2554.90.00faibleBon signalAcquisitionmean(abs(X)) / noise_rmsalert decreases with SNR dB; >=40 dB is treated as low alert
BruitBandwise SNRnoise.bandwise_snr_min523.040.00faibleZone 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_count650.18faibleSpectre propreCosmic rays, splicecount robust outliers in second derivativealert follows spike_rate, saturated at 1%
Artefacts locauxSpike rateartefacts.spike_rate0.18%0.18faibleNormalInterpolationspike_count / (n_samples * (n_features - 2))alert = min(1, spike_rate / 0.01)
Artefacts locauxJump countartefacts.jump_count620.17faibleContinuSplicecount robust outliers in first derivativealert follows jump_rate, saturated at 1%
Artefacts locauxJump rateartefacts.jump_rate0.172%0.17faibleNormalCalibrationjump_count / (n_samples * (n_features - 1))alert = min(1, jump_rate / 0.01)
Artefacts locauxClip fractionartefacts.clip_fraction0.00554%0.01faibleNormalDétecteur saturéfraction of finite cells equal to repeated min/max extremaalert = min(1, clip_fraction / 0.01)
Forme spectraleBaseline slopeshape.baseline_slope-2.53240.06faibleStableÉclairementlinear slope of mean_spectrum over normalized axisalert = abs(slope / signal_scale), saturated at 0.5
Forme spectraleCurvature RMSshape.curvature_rms0.118880.13faibleLisseFond, splicemedian RMS(second derivative per spectrum)alert = curvature_rms / signal_scale, saturated at 1%
Forme spectraleD1 RMSshape.d1_rms0.094070.02faiblePlatBiologie ou artefactmedian RMS(first derivative per spectrum)alert = d1_rms / signal_scale, saturated at 5%
Outliers multivariésPCA Q (SPE)outliers.pca_q_ratio8.49541.00fortSpectre atypiqueArtefact, mélangep95(Q/SPE residual) / median(Q/SPE residual)alert = min(1, pca_q_ratio / 8)
Outliers multivariésHotelling T²outliers.hotelling_t2_ratio2.66910.33faibleCentralVariabilité naturellep95(Hotelling T2) / median(Hotelling T2)alert = min(1, hotelling_t2_ratio / 8)
Outliers multivariésMahalanobis Houtliers.mahalanobis_h_ratio1.62240.41faiblePopulation normaleDomaine différentp95(sqrt(T2)) / median(sqrt(T2))alert = min(1, mahalanobis_h_ratio / 4)
Comparaison à référenceRMS to mean spectrumreference.rms_to_mean_spectrum_p953.6020.16faibleTypiqueDomain 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_p950.00860840.02faibleSimilaireFond, 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_density0.0162240.75fortSous-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_p952.26070.63moyenSpectre 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_p950.597840.75fortSpectre atypiqueDiverses causesp95 IsolationForest anomaly score on PCA scoresalert follows structure complexity; raw score is implementation-dependent
X PCA score plot-200-1000100200300-40-2002040PC1 -126.5 · PC2 25.56PC1 87.78 · PC2 -39.06PC1 -12.89 · PC2 -24.29PC1 -142.5 · PC2 -19.38PC1 -42.5 · PC2 -23.29PC1 -119.8 · PC2 -20.38PC1 103.4 · PC2 10.06PC1 252.2 · PC2 14.67PC1 -41.62 · PC2 26.84PC1 -72.31 · PC2 24.64PC1 -36.95 · PC2 21.45PC1 138.8 · PC2 2.384PC1 -4.669 · PC2 12.3PC1 -35.97 · PC2 14.81PC1 28.67 · PC2 5.64PC1 24.89 · PC2 -31.95PC1 (94.9%)PC2 (4.4%)16 scores
PCA explained variance0%25%50%75%100%PC1: 94.9% (cumulative 94.9%)1PC2: 4.4% (cumulative 99.3%)2PC3: 0.5% (cumulative 99.8%)3PC4: 0.1% (cumulative 99.9%)4PC5: 0.0% (cumulative 99.9%)5PC6: 0.0% (cumulative 99.9%)6PC7: 0.0% (cumulative 100.0%)7PC8: 0.0% (cumulative 100.0%)8PC9: 0.0% (cumulative 100.0%)9PC10: 0.0% (cumulative 100.0%)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 1

material_name

target · categorical
material_name classesBrass plateBrass plate: 1515Gold plateGold plate: 11
n / missing16 / 0
Classes2
Balance (entropy)0.34
Imbalance ratio15
Top classBrass plate (15)

Metadata 2

ecostress_resource_id

metadata · categorical
n / missing16 / 0
Classes16
Balance (entropy)1
Imbalance ratio1
Top classmanmade.reflectancetarget.none.solid.tir.001682.jpl.nicolet.spectrum (1)

sample_description

metadata · categorical
n / missing16 / 0
Classes16
Balance (entropy)1
Imbalance ratio1
Top classMeasurement made at center. Field Gold.Oct. 29 1997 to Oct. 30 1997. Original ASTER Spectral Library name was jpl.nicolet.manmade.target.none.solid.fg11097j.spectrum.txt (1)
Constant metadata 16
  • categorymanmade
  • material_typeManmade
  • locationLabsphere, P.O. Box 70, Shaker Street, North Sutton, NH 03260(603) 927-4266
  • instrumentjpl.nicolet
  • acquisition_modeHemispherical Reflectance
  • signal_typeReflectence (percent)
  • axis_unitWavelength (micrometers)
  • axis_min2
  • axis_max15.39
  • n_points_original2,256
  • 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
Samples16
Observations (total)16
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 hashf64f200fe6ec25f5…
Processing hashee5c1faa67913ced…
Metadata hash11b718918de202dc…

Load this dataset

# pip install nirs4all-datasets
from nirs4all_datasets import get

# private dataset — export requires a Dataverse token
ds = get("ecostress_manmade_tir_2256points", 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.