← Back to the catalog
Private

ECOSTRESS mineral tir axis d3032c60

ecostress · other

ECOSTRESS mineral tir axis d3032c60. v2.0 standardized NIRS package: 1 spectral source(s), 2 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.
5
samples
2,287
wavelengths
1
sources
2
targets
27
metadata
other
family

Dataset property explorer

Mean profile risk0.49
Highest axisArtefacts locaux · 1.00
Diagnostics8
Sources profiled1
ECOSTRESS mineral tir axis d3032c60 property profile0.250.50.751integritynoiseartefactsbaselinePCA outliersreferencerepeatabilitystructureECOSTRESS mineral tir axis d3032c60 profileintegrity: 0.00noise: 0.00artefacts: 1.00baseline: 1.00PCA outliers: 0.51reference: 1.00repeatability: 0.00structure: 0.39ECOSTRESS miner…0 center · 1 outer ring · outward = stronger anomaly / heterogeneity signal

Profile axes

Intégrité0.00
Artefacts locaux1.00
Bruit0.00
Outliers PCA0.51
Distance à la référence1.00
Répétabilité0.00
Baseline / forme1.00
Structure multi-régimes0.39
Diagnostic hypotheses00.250.50.751hypothesis scoreSplice / raccord détecteursSplice / raccord détecteurs: 0.840.84Erreur calibration / référenc…Erreur calibration / référence blanche: 0.710.71Erreur interpolation / réécha…Erreur interpolation / rééchantillonnage: 0.640.64Fond différentFond différent: 0.630.63Signature VERA25-likeSignature VERA25-like: 0.600.60Différence de sonde / géométr…Différence de sonde / géométrie: 0.550.55Mélange feuille + fondMélange feuille + fond: 0.460.46Spectre saturé / clippingSpectre saturé / clipping: 0.430.43
DiagnosticScoreForceSignauxInterprétation probable
Splice / raccord détecteursX0.84forteSpike rate 1.00, Jump rate 1.00, RMS/SAM référence 1.00Rupture aux jonctions de détecteurs, calibration locale ou sonde différente.
Erreur calibration / référence blancheX0.71moyenneBaseline/mean/area 1.00, RMS/SAM référence 1.00, artefacts locaux 1.00Décalage systématique entre campagnes, instruments ou référence blanche.
Erreur interpolation / rééchantillonnageX0.64moyenneSpike rate 1.00, Jump rate 1.00, SNR normal/élevé 1.00Artefacts numériques ou traitement spectral incorrect.
Fond différentX0.63moyenneBaseline/mean/area 1.00, RMS/SAM référence 1.00, PCA Q 0.51Effet systématique du support, blanc/noir, transflectance ou environnement de mesure.
Signature VERA25-likeX0.60moyenneSpike rate 1.00, Jump rate 1.00, RMS/SAM référence 1.00Combinaison possible changement de sonde + splice, amplifiée par géométrie, fond ou calibration.
Différence de sonde / géométrieX0.55moyenneBaseline/mean/area 1.00, RMS/SAM référence 1.00, PCA Q 0.51Modification de l'illumination, collecte, angle ou distance sonde-échantillon.
Mélange feuille + fondX0.46moyenneBaseline/mean/area 1.00, RMS/SAM référence 1.00, PCA Q 0.51Couverture partielle du spot; contribution du fond ou du support.
Spectre saturé / clippingX0.43moyenneBaseline/mean/area 1.00, Jump rate 1.00, PCA Q 0.51Détecteur saturé ou dynamique insuffisante.

Spectral sources

mineral tir

X · other · source instruments vary by sample
mineral tir spectra02550751001250102030q05-q95 envelopeq25-q75 envelopemedian spectrummedianq25–q75q05–q95wavelength / none25.044none — median 48.87 (q25–q75 47.14–60.3)23.247none — median 46.6 (q25–q75 40.45–50.7)21.601none — median 37.72 (q25–q75 33.81–39.06)20.251none — median 50.48 (q25–q75 36.4–58.03)18.99none — median 53.52 (q25–q75 35.89–66.9)17.938none — median 64.36 (q25–q75 58.29–71.43)16.942none — median 68.17 (q25–q75 56.58–85.37)16.1none — median 66.05 (q25–q75 53.7–78.05)15.292none — median 64.73 (q25–q75 53.84–72.12)14.603none — median 71.21 (q25–q75 55.81–71.64)13.973none — median 71.09 (q25–q75 65.83–76.38)13.361none — median 73.29 (q25–q75 72.39–79.21)12.832none — median 74.51 (q25–q75 73.56–81.8)12.314none — median 74.2 (q25–q75 73.47–77.84)11.863none — median 70.07 (q25–q75 46.36–79.88)11.419none — median 60.07 (q25–q75 29.32–70.33)11.03none — median 57.67 (q25–q75 43.56–66.3)10.645none — median 57.33 (q25–q75 46.81–60.44)10.306none — median 28.2 (q25–q75 25.24–42.7)9.9887none — median 35.15 (q25–q75 24.46–42.33)9.6719none — median 34.13 (q25–q75 12.16–58.6)9.3916none — median 26.6 (q25–q75 11–42.36)9.111none — median 22.4 (q25–q75 17.21–28.65)8.8618none — median 25.24 (q25–q75 20.13–59.71)8.6116none — median 58.31 (q25–q75 31.25–74.43)8.3886none — median 74.8 (q25–q75 41.31–82.21)8.164none — median 85.08 (q25–q75 64.17–88.05)7.9634none — median 89.82 (q25–q75 74.82–90.8)7.7723none — median 92.46 (q25–q75 81.05–92.58)7.5792none — median 93.85 (q25–q75 84.04–94.77)7.4059none — median 95.1 (q25–q75 85.7–95.9)7.2303none — median 95.96 (q25–q75 86.4–96.83)7.0725none — median 96.51 (q25–q75 86.75–97.21)6.9122none — median 96.98 (q25–q75 87.44–97.37)6.7678none — median 97.34 (q25–q75 88–97.64)6.6209none — median 97.34 (q25–q75 88.89–97.6)6.4883none — median 97.02 (q25–q75 89.1–97.43)6.3531none — median 97.26 (q25–q75 89.05–97.38)6.231none — median 95.88 (q25–q75 87.74–97.25)6.1134none — median 93.71 (q25–q75 81.51–97.49)5.9932none — median 96.03 (q25–q75 84.88–96.71)5.8844none — median 97.5 (q25–q75 88.13–97.96)5.773none — median 97.93 (q25–q75 89.61–98.4)5.6719none — median 97.91 (q25–q75 90.12–98.76)5.5684none — median 98.06 (q25–q75 90.22–98.99)5.4743none — median 98.46 (q25–q75 90.38–99.06)5.3778none — median 98.62 (q25–q75 90.52–99.23)5.29none — median 98.55 (q25–q75 90.78–99.44)5.205none — median 98.4 (q25–q75 90.94–99.64)5.1176none — median 98.48 (q25–q75 91.13–99.5)5.038none — median 98.21 (q25–q75 91.23–99.35)4.9562none — median 97.69 (q25–q75 91.35–99.51)4.8815none — median 97.22 (q25–q75 91.39–99.58)4.8046none — median 97.57 (q25–q75 91.4–99.62)4.7344none — median 97.99 (q25–q75 91.33–99.67)4.662none — median 97.9 (q25–q75 91.25–99.62)4.5959none — median 97.81 (q25–q75 91.28–99.66)4.5316none — median 97.81 (q25–q75 91.22–99.53)4.4652none — median 97.71 (q25–q75 91.2–99.37)4.4045none — median 97.56 (q25–q75 91.13–99.27)4.3418none — median 97.34 (q25–q75 91.2–99.16)4.2844none — median 97.17 (q25–q75 91.13–99.11)4.2251none — median 96.94 (q25–q75 91.22–99.05)4.1707none — median 96.73 (q25–q75 91.23–98.9)4.1144none — median 96.53 (q25–q75 91.23–98.8)4.0628none — median 96.28 (q25–q75 91.22–98.73)4.0125none — median 96.04 (q25–q75 91.14–98.61)3.9604none — median 95.75 (q25–q75 91.06–98.42)3.9126none — median 95.53 (q25–q75 91.09–98.31)3.863none — median 95.12 (q25–q75 90.94–98.14)3.8175none — median 94.84 (q25–q75 90.84–98.02)3.7703none — median 94.41 (q25–q75 90.74–97.9)3.7269none — median 94.09 (q25–q75 90.45–97.65)3.6819none — median 93.6 (q25–q75 90.15–97.41)3.6406none — median 93.24 (q25–q75 89.84–97.31)3.6001none — median 92.61 (q25–q75 89.4–97.04)3.5581none — median 92.29 (q25–q75 89–96.8)3.5195none — median 92.03 (q25–q75 88.33–96.73)3.4793none — median 91.47 (q25–q75 87.55–96.38)3.4423none — median 91.04 (q25–q75 86.79–96.3)3.4039none — median 90.23 (q25–q75 85.63–96.03)3.3685none — median 89.25 (q25–q75 85.13–95.65)3.3317none — median 88.28 (q25–q75 84.2–95.12)3.2978none — median 87.33 (q25–q75 83.07–94.85)3.2646none — median 86.4 (q25–q75 81.67–94.57)3.23none — median 85.24 (q25–q75 79.89–94.15)3.1981none — median 83.79 (q25–q75 78.21–93.63)3.1649none — median 82.34 (q25–q75 76.28–93.34)3.1343none — median 80.1 (q25–q75 74.37–93.09)3.1024none — median 77.38 (q25–q75 72.36–92.79)3.073none — median 74.7 (q25–q75 70.92–92.68)3.0423none — median 70.75 (q25–q75 69.76–90.45)3.014none — median 68.69 (q25–q75 58.37–88.02)2.9863none — median 67.98 (q25–q75 49.39–86.15)2.9573none — median 67.62 (q25–q75 58.18–83.75)2.9306none — median 69.44 (q25–q75 67.71–81.74)2.9027none — median 71.15 (q25–q75 67.98–82.02)2.8769none — median 74.42 (q25–q75 67.98–84.06)2.85none — median 77.12 (q25–q75 68.26–83.52)2.8252none — median 78.51 (q25–q75 68.79–80.56)2.7992none — median 80.02 (q25–q75 69.51–80.74)2.7752none — median 83.2 (q25–q75 70.23–89.22)2.7517none — median 85.78 (q25–q75 72.9–92.75)2.7271none — median 87.18 (q25–q75 78.39–93.36)2.7043none — median 87.45 (q25–q75 81.74–93.26)2.6805none — median 87.39 (q25–q75 83.17–93.33)2.6585none — median 87.12 (q25–q75 83.45–93.22)2.6356none — median 86.68 (q25–q75 83.76–93.1)2.6143none — median 86.27 (q25–q75 83.81–92.89)2.5921none — median 85.9 (q25–q75 83.92–92.68)2.5715none — median 85.59 (q25–q75 83.53–92.75)2.55none — median 85.3 (q25–q75 83.5–92.41)2.5301none — median 84.84 (q25–q75 83.57–92.19)2.5105none — median 84.62 (q25–q75 83.33–92.19)2.49none — median 84.25 (q25–q75 83.33–91.94)2.471none — median 83.82 (q25–q75 83.02–91.79)2.4511none — median 83.57 (q25–q75 82.94–91.71)2.4327none — median 83.09 (q25–q75 83.08–91.47)2.4135none — median 82.93 (q25–q75 82.78–91.34)2.3956none — median 82.88 (q25–q75 82.36–91.2)2.377none — median 82.76 (q25–q75 82.06–91.02)2.3597none — median 82.37 (q25–q75 81.52–90.69)2.3426none — median 82.54 (q25–q75 81.32–90.66)2.3247none — median 82.28 (q25–q75 81.02–90.28)2.3082none — median 82.43 (q25–q75 81.12–90.08)2.2908none — median 82.12 (q25–q75 80.85–90.23)2.2747none — median 81.96 (q25–q75 80.81–89.81)2.2579none — median 81.99 (q25–q75 80.75–89.76)2.2423none — median 81.88 (q25–q75 80.78–89.59)2.2259none — median 82.01 (q25–q75 80.55–89.07)2.2107none — median 81.87 (q25–q75 80.8–89.1)2.1957none — median 81.9 (q25–q75 80.25–89.09)2.18none — median 81.72 (q25–q75 80.23–88.68)2.1655none — median 82.25 (q25–q75 80.57–88.58)2.1502none — median 82.03 (q25–q75 80.03–88.02)2.136none — median 81.87 (q25–q75 80.26–88.11)2.1212none — median 82.2 (q25–q75 80.2–87.74)2.1074none — median 82.21 (q25–q75 80.33–87.58)2.0929none — median 82.15 (q25–q75 80.26–87.57)2.0795none — median 82.28 (q25–q75 79.98–87.3)

Sampling

Wavelengths2,287
Axis range2.079–25.04 none
Mean spacing0.01 none
Gridirregular
Observations5

Signal & quality

Value range3.45 – 107
Mean range28.6 – 95.9
Mean level82.39
Area1457
PTP67.26
Noise RMS0.020577
SNR4e+03
SNR dB7e+01 dB
Dynamic range67.3
Smoothness0.1257
Saturated0.0%
X-outliers1

Integrity & artefacts

NaN ratio0.00%
Inf count0
Zero ratio0.00%
Spike count219
Spike rate1.92%
Jump count1,011
Jump rate8.85%
Clip fraction0.02%

Shape & reference

Baseline slope-54.818
Curvature RMS0.10294
D1 RMS0.31088
RMS to mean10.962
RMS p9513.993
SAM to mean0.10067
SAM p950.12417
Affine offset p9535.315
Affine gain p95 Δ0.33825
Affine residual p958.3487
Xcorr lag p9512

Outliers & repeatability

PCA Q p95/median4.1
Hotelling T2 p95/median1.1
Mahalanobis H p95/median1.1
Repeat groups0

Dimensionality (PCA)

Effective rank2.5
PCs → 95% var3
PCs → 99% var4
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_reflectance82.3861.00fortValeur atypique: Trop clair / fond visible ou Trop sombreFond, géométriemean(X finite)alert reuses baseline/shape drift because absolute reflectance ranges are technology-dependent
Amplitude globaleArea under curveamplitude.area_under_curve1457.51.00fortValeur atypique: Différence d'éclairement ou NormalDistance 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_peak67.2560.00faibleVariabilité forteSaturationmax(mean_spectrum) - min(mean_spectrum)alert increases when dynamic range is abnormally flat
Amplitude globaleVarianceamplitude.variance390.950.00faibleNormal ou hétérogèneMauvais contactvar(X finite)alert increases when variance/dynamic range is abnormally flat
BruitNoise RMSnoise.noise_rms0.0205770.00faibleStableLampe, détecteurmedian MAD(second derivative) * 1.4826 / sqrt(6)alert = noise_rms / signal_scale, saturated at 5%
BruitSNRnoise.snr4003.80.00faibleBon signalAcquisitionmean(abs(X)) / noise_rmsalert decreases with SNR dB; >=40 dB is treated as low alert
BruitBandwise SNRnoise.bandwise_snr_min145.840.00faibleZone 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_count2191.00fortArtefactsCosmic rays, splicecount robust outliers in second derivativealert follows spike_rate, saturated at 1%
Artefacts locauxSpike rateartefacts.spike_rate1.92%1.00fortSpectre suspectInterpolationspike_count / (n_samples * (n_features - 2))alert = min(1, spike_rate / 0.01)
Artefacts locauxJump countartefacts.jump_count1,0111.00fortRaccord détecteurSplicecount robust outliers in first derivativealert follows jump_rate, saturated at 1%
Artefacts locauxJump rateartefacts.jump_rate8.85%1.00fortProblème spectralCalibrationjump_count / (n_samples * (n_features - 1))alert = min(1, jump_rate / 0.01)
Artefacts locauxClip fractionartefacts.clip_fraction0.0175%0.02faibleNormalDétecteur saturéfraction of finite cells equal to repeated min/max extremaalert = min(1, clip_fraction / 0.01)
Forme spectraleBaseline slopeshape.baseline_slope-54.8181.00fortDériveÉclairementlinear slope of mean_spectrum over normalized axisalert = abs(slope / signal_scale), saturated at 0.5
Forme spectraleCurvature RMSshape.curvature_rms0.102940.12faibleLisseFond, splicemedian RMS(second derivative per spectrum)alert = curvature_rms / signal_scale, saturated at 1%
Forme spectraleD1 RMSshape.d1_rms0.310880.08faiblePlatBiologie ou artefactmedian RMS(first derivative per spectrum)alert = d1_rms / signal_scale, saturated at 5%
Outliers multivariésPCA Q (SPE)outliers.pca_q_ratio4.0720.51moyenSpectre atypiqueArtefact, mélangep95(Q/SPE residual) / median(Q/SPE residual)alert = min(1, pca_q_ratio / 8)
Outliers multivariésHotelling T²outliers.hotelling_t2_ratio1.1390.14faibleCentralVariabilité naturellep95(Hotelling T2) / median(Hotelling T2)alert = min(1, hotelling_t2_ratio / 8)
Outliers multivariésMahalanobis Houtliers.mahalanobis_h_ratio1.06710.27faiblePopulation normaleDomaine différentp95(sqrt(T2)) / median(sqrt(T2))alert = min(1, mahalanobis_h_ratio / 4)
Comparaison à référenceRMS to mean spectrumreference.rms_to_mean_spectrum_p9513.9930.68moyenSpectre 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_p950.124170.35faibleSimilaireFond, 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_density0.0012440.39faibleHomogèneLots 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_p951.16160.08faiblePopulation normaleCas raresp95 approximate LOF from PCA-score kNN distancesalert = min(1, max(0, LOF_p95 - 1) / 2)
Structure du datasetIsolation Forest scorestructure.isolation_forest_score_p950.504080.39faibleNormalDiverses causesp95 IsolationForest anomaly score on PCA scoresalert follows structure complexity; raw score is implementation-dependent
X PCA score plot-1,000-50005001,000-400-2000200400600PC1 -347.3 · PC2 166.7PC1 -641.2 · PC2 -184.3PC1 220 · PC2 454.5PC1 519.3 · PC2 -190.6PC1 249.1 · PC2 -246.2PC1 (62.9%)PC2 (25.2%)5 scores
PCA explained variance0%25%50%75%100%PC1: 62.9% (cumulative 62.9%)1PC2: 25.2% (cumulative 88.1%)2PC3: 10.4% (cumulative 98.6%)3PC4: 1.4% (cumulative 100.0%)4cumulative 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 2

material_name

target · categorical
n / missing5 / 0
Classes5
Balance (entropy)1
Imbalance ratio1
Top classDiopside CaMgSi2O6 (1)

subclass

target · categorical
subclass classesTectosilicateTectosilicate: 22InosilicateInosilicate: 11PhyllosilicatePhyllosilicate: 11SorosilicateSorosilicate: 11
n / missing5 / 0
Classes4
Balance (entropy)0.96
Imbalance ratio2
Top classTectosilicate (2)

Metadata 4

ecostress_resource_id

metadata · categorical
n / missing5 / 0
Classes5
Balance (entropy)1
Imbalance ratio1
Top classmineral.silicate.inosilicate.fine.tir.diopside_2.jhu.nicolet.spectrum (1)

location

metadata · categorical
n / missing5 / 0
Classes5
Balance (entropy)1
Imbalance ratio1
Top classSample from DeKalb, New York via the Smithsonian (sampleno. NMNH R18685). (1)

sample_description

metadata · categorical
n / missing5 / 0
Classes5
Balance (entropy)1
Imbalance ratio1
Top classMost of the 13 mm x 10 mm x 6 mm sample was clean, fresh and green. One area was eroded and weathered and probably impure. The crushed sample was hand-picked to avoid impurities. A moderate amount of weathering and alteration can be seen under the microscope. The alteration is brown in color (limonite). Particle size was less than 2 Micrometers.(Chain Silicates) (Pyroxene Group) Original ASTER Spectral Library name was jhu.nicolet.mineral.silicate.inosilicate.powder.diopsi2.spectrum.txt (1)

notes

metadata · categorical
n / missing5 / 1
Classes4
Balance (entropy)1
Imbalance ratio1
Top classmineral.silicate.inosilicate.fine.tir.diopside_2.jhu.nicolet.ancillary.txt (1)
Constant metadata 14
  • categorymineral
  • material_typeMineral
  • instrumentjhu.nicolet
  • acquisition_modeTransmission
  • signal_typeTransmittance (percent)
  • axis_unitWavelength (micrometers)
  • axis_min2.079
  • axis_max25.04
  • n_points_original2,287
  • 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

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

Alignment

Alignment levelobservation
Sample id availableyes
Samples5
Observations (total)5
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 hash5318c567d832083c…
Processing hash3938dd62e3fbc6d6…
Metadata hash27d259e5665045f5…

Load this dataset

# pip install nirs4all-datasets
from nirs4all_datasets import get

# private dataset — export requires a Dataverse token
ds = get("ecostress_mineral_tir_2287points_5rows", 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.