← Back to the catalog
Privatetoken required

GitHub grapevine chloride SVC HR-1024i VIS-NIR

github · NIR

GitHub grapevine chloride SVC HR-1024i VIS-NIR. v2.0 standardized NIRS package: 1 spectral source(s), 4 declared target(s). Auto-generated from dataset_card.json (verify before publication).

nirv2github
🔒
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.
556
samples
—
wavelengths
1
sources
4
targets
20
metadata
NIR
family

Dataset property explorer

Mean profile risk0.43
Highest axisArtefacts locaux · 1.00
Diagnostics8
Sources profiled1
GitHub grapevine chloride SVC HR-1024i VIS-NIR property profile0.250.50.751integritynoiseartefactsbaselinePCA outliersreferencerepeatabilitystructureGitHub grapevine chloride SVC HR-1024i VIS-NIR profileintegrity: 0.00noise: 0.00artefacts: 1.00baseline: 0.68PCA outliers: 0.48reference: 0.40repeatability: 0.00structure: 0.84GitHub grapevin…0 center · 1 outer ring · outward = stronger anomaly / heterogeneity signal

Profile axes

Intégrité0.00
Artefacts locaux1.00
Bruit0.00
Outliers PCA0.48
Distance à la référence0.40
Répétabilité0.00
Baseline / forme0.68
Structure multi-régimes0.84
Diagnostic hypotheses00.250.50.751hypothesis scoreSplice / raccord détecteursSplice / raccord détecteurs: 0.680.68Erreur interpolation / réécha…Erreur interpolation / rééchantillonnage: 0.580.58Signature VERA25-likeSignature VERA25-like: 0.510.51Erreur calibration / référenc…Erreur calibration / référence blanche: 0.490.49Fond différentFond différent: 0.420.42Différence de sonde / géométr…Différence de sonde / géométrie: 0.420.42Spectre hors domaine valideSpectre hors domaine valide: 0.390.39Dataset multi-régimesDataset multi-régimes: 0.370.37
DiagnosticScoreForceSignauxInterprétation probable
Splice / raccord détecteursX0.68moyenneSpike rate 1.00, Jump rate 1.00, SNR non dégradé 1.00Rupture aux jonctions de détecteurs, calibration locale ou sonde différente.
Erreur interpolation / rééchantillonnageX0.58moyenneSpike rate 1.00, Jump rate 1.00, SNR normal/élevé 1.00Artefacts numériques ou traitement spectral incorrect.
Signature VERA25-likeX0.51moyenneSpike rate 1.00, Jump rate 1.00, Mahalanobis / T2 0.48Combinaison possible changement de sonde + splice, amplifiée par géométrie, fond ou calibration.
Erreur calibration / référence blancheX0.49moyenneartefacts locaux 1.00, Baseline/mean/area 0.68, Mahalanobis / T2 0.48Décalage systématique entre campagnes, instruments ou référence blanche.
Fond différentX0.42moyenneBaseline/mean/area 0.68, Mahalanobis / T2 0.48, RMS/SAM référence 0.40Effet systématique du support, blanc/noir, transflectance ou environnement de mesure.
Différence de sonde / géométrieX0.42faibleBaseline/mean/area 0.68, Mahalanobis / T2 0.48, RMS/SAM référence 0.40Modification de l'illumination, collecte, angle ou distance sonde-échantillon.
Spectre hors domaine valideX0.39faibleStructure PCA 0.84, Mahalanobis / T2 0.48, RMS/SAM référence 0.40Variété, espèce, lot ou condition différente mais physiquement plausible.
Dataset multi-régimesX0.37faibleStructure PCA 0.84, Mahalanobis / T2 0.48, RMS/SAM référence 0.40Mélange de campagnes, opérateurs, lots, setups ou sous-populations spectrales.

Spectral sources

X

X · NIR · SVC HR-1024i
X spectra020406001,0002,0003,000q05-q95 envelopeq25-q75 envelopemedian spectrummedianq25–q75q05–q95wavelength / nm339nm — median 9.94 (q25–q75 7.665–12.22)355nm — median 6.562 (q25–q75 5.425–7.711)370nm — median 5.012 (q25–q75 4.309–5.671)386nm — median 4.404 (q25–q75 3.905–4.993)402nm — median 4.087 (q25–q75 3.655–4.609)417nm — median 4.017 (q25–q75 3.574–4.461)433nm — median 4.032 (q25–q75 3.579–4.447)449nm — median 4.044 (q25–q75 3.588–4.494)464nm — median 4.12 (q25–q75 3.683–4.561)480nm — median 4.205 (q25–q75 3.75–4.651)496nm — median 4.429 (q25–q75 3.967–4.905)511nm — median 5.305 (q25–q75 4.739–5.979)527nm — median 8.518 (q25–q75 7.352–10.21)543nm — median 10.57 (q25–q75 9.024–12.73)558nm — median 10.74 (q25–q75 9.156–12.95)574nm — median 8.412 (q25–q75 7.156–10.2)589nm — median 6.982 (q25–q75 6.031–8.361)605nm — median 6.434 (q25–q75 5.598–7.631)621nm — median 5.574 (q25–q75 4.971–6.497)636nm — median 5.331 (q25–q75 4.757–6.157)652nm — median 4.692 (q25–q75 4.212–5.239)668nm — median 4.379 (q25–q75 3.928–4.821)683nm — median 4.802 (q25–q75 4.346–5.328)699nm — median 9.764 (q25–q75 8.492–11.53)715nm — median 24.3 (q25–q75 21.78–27.37)730nm — median 37.43 (q25–q75 35.46–39.83)746nm — median 44.76 (q25–q75 43–46.83)762nm — median 46.71 (q25–q75 44.9–48.93)777nm — median 47.03 (q25–q75 45.24–49.26)793nm — median 47.18 (q25–q75 45.37–49.37)809nm — median 47.28 (q25–q75 45.42–49.41)824nm — median 47.31 (q25–q75 45.47–49.45)840nm — median 47.34 (q25–q75 45.49–49.48)856nm — median 47.37 (q25–q75 45.51–49.55)871nm — median 47.38 (q25–q75 45.53–49.55)887nm — median 47.32 (q25–q75 45.45–49.55)903nm — median 47.28 (q25–q75 45.42–49.53)918nm — median 47.24 (q25–q75 45.41–49.46)934nm — median 47.06 (q25–q75 45.2–49.27)950nm — median 46.67 (q25–q75 44.79–48.88)965nm — median 46.15 (q25–q75 44.33–48.33)981nm — median 46.08 (q25–q75 44.23–48.24)996nm — median 46.06 (q25–q75 44.31–48.22)1,012nm — median 45.86 (q25–q75 44.09–48.09)1,028nm — median 45.99 (q25–q75 44.33–48.23)1,043nm — median 46.16 (q25–q75 44.43–48.31)1,059nm — median 46.26 (q25–q75 44.49–48.39)1,075nm — median 46.27 (q25–q75 44.49–48.41)1,090nm — median 46.24 (q25–q75 44.46–48.41)1,106nm — median 46.07 (q25–q75 44.27–48.23)1,122nm — median 45.8 (q25–q75 44.03–47.97)1,137nm — median 45.09 (q25–q75 43.36–47.24)1,153nm — median 43.66 (q25–q75 41.91–45.75)1,169nm — median 43.05 (q25–q75 41.43–45.13)1,184nm — median 42.89 (q25–q75 41.29–44.93)1,200nm — median 42.81 (q25–q75 41.26–44.86)1,216nm — median 42.94 (q25–q75 41.35–45.01)1,231nm — median 43.1 (q25–q75 41.51–45.21)1,247nm — median 43.31 (q25–q75 41.71–45.41)1,263nm — median 43.33 (q25–q75 41.72–45.48)1,278nm — median 43.22 (q25–q75 41.6–45.35)1,294nm — median 42.88 (q25–q75 41.26–45.01)1,310nm — median 42.13 (q25–q75 40.52–44.22)1,325nm — median 40.92 (q25–q75 39.19–42.88)1,341nm — median 39.23 (q25–q75 37.43–41.15)1,357nm — median 37.49 (q25–q75 35.73–39.29)1,372nm — median 35.18 (q25–q75 33.56–36.97)1,388nm — median 29.14 (q25–q75 27.61–30.87)1,404nm — median 20.82 (q25–q75 19.34–22.45)1,419nm — median 16.5 (q25–q75 15.03–17.91)1,435nm — median 14.93 (q25–q75 13.52–16.31)1,450nm — median 14.6 (q25–q75 13.28–16.15)1,466nm — median 15.12 (q25–q75 13.75–16.75)1,482nm — median 16.65 (q25–q75 15.2–18.39)1,497nm — median 18.62 (q25–q75 17.07–20.34)1,513nm — median 20.78 (q25–q75 19.12–22.56)1,529nm — median 22.87 (q25–q75 21.17–24.62)1,544nm — median 24.62 (q25–q75 22.95–26.37)1,560nm — median 26.21 (q25–q75 24.61–27.92)1,576nm — median 27.56 (q25–q75 25.96–29.2)1,591nm — median 28.62 (q25–q75 27.04–30.27)1,607nm — median 29.48 (q25–q75 27.92–31.16)1,623nm — median 30.26 (q25–q75 28.62–31.92)1,638nm — median 30.78 (q25–q75 29.15–32.42)1,654nm — median 31.14 (q25–q75 29.46–32.79)1,670nm — median 31.2 (q25–q75 29.53–32.85)1,685nm — median 31.01 (q25–q75 29.4–32.71)1,701nm — median 30.5 (q25–q75 28.97–32.16)1,717nm — median 29.9 (q25–q75 28.33–31.51)1,732nm — median 29.2 (q25–q75 27.61–30.75)1,748nm — median 28.35 (q25–q75 26.74–29.88)1,764nm — median 27.3 (q25–q75 25.73–28.85)1,779nm — median 26.63 (q25–q75 25.11–28.14)1,795nm — median 26.36 (q25–q75 24.85–27.89)1,811nm — median 26.35 (q25–q75 24.81–27.85)1,826nm — median 26.26 (q25–q75 24.75–27.78)1,842nm — median 25.56 (q25–q75 24.04–27.08)1,858nm — median 23.12 (q25–q75 21.74–24.76)1,873nm — median 17.83 (q25–q75 16.43–19.23)1,889nm — median 9.934 (q25–q75 8.934–11.1)1,904nm — median 5.825 (q25–q75 5.218–6.549)1,920nm — median 3.789 (q25–q75 3.353–4.343)1,936nm — median 3.807 (q25–q75 3.35–4.351)1,951nm — median 4.122 (q25–q75 3.606–4.695)1,967nm — median 4.705 (q25–q75 4.082–5.395)1,983nm — median 5.51 (q25–q75 4.748–6.311)1,998nm — median 6.413 (q25–q75 5.56–7.327)2,014nm — median 7.498 (q25–q75 6.548–8.539)2,030nm — median 8.572 (q25–q75 7.511–9.753)2,045nm — median 9.512 (q25–q75 8.388–10.76)2,061nm — median 10.43 (q25–q75 9.303–11.76)2,077nm — median 11.46 (q25–q75 10.28–12.76)2,092nm — median 12.36 (q25–q75 11.11–13.76)2,108nm — median 13.29 (q25–q75 11.98–14.68)2,124nm — median 14.14 (q25–q75 12.82–15.61)2,139nm — median 14.79 (q25–q75 13.46–16.26)2,155nm — median 15.3 (q25–q75 13.94–16.76)2,171nm — median 15.63 (q25–q75 14.23–17.06)2,186nm — median 15.95 (q25–q75 14.48–17.35)2,202nm — median 16.22 (q25–q75 14.74–17.58)2,218nm — median 16.33 (q25–q75 14.88–17.69)2,233nm — median 16.18 (q25–q75 14.75–17.6)2,249nm — median 15.67 (q25–q75 14.26–17.01)2,265nm — median 14.78 (q25–q75 13.41–16.11)2,280nm — median 14 (q25–q75 12.67–15.27)2,296nm — median 13.25 (q25–q75 11.95–14.52)2,311nm — median 12.46 (q25–q75 11.23–13.72)2,327nm — median 11.77 (q25–q75 10.57–13.04)2,343nm — median 11.04 (q25–q75 9.884–12.26)2,358nm — median 10.33 (q25–q75 9.268–11.51)2,374nm — median 9.614 (q25–q75 8.57–10.77)2,390nm — median 8.885 (q25–q75 7.902–9.962)2,405nm — median 8.214 (q25–q75 7.272–9.205)2,421nm — median 7.477 (q25–q75 6.575–8.385)2,437nm — median 6.744 (q25–q75 5.938–7.595)2,452nm — median 6.142 (q25–q75 5.441–6.946)2,468nm — median 5.605 (q25–q75 4.92–6.361)2,484nm — median 5.2 (q25–q75 4.558–5.87)2,499nm — median 4.899 (q25–q75 4.288–5.561)2,515nm — median 4.985 (q25–q75 4.269–5.732)

Sampling

Wavelengths2,177
Axis range339–2,515 nm
Mean spacing1 nm
Griduniform
Observations556

Signal & quality

Value range-0.355 – 68.2
Mean range3.86 – 47.7
Mean level23.52
Area5.119e+04
PTP43.85
Noise RMS0.0025609
SNR9.2e+03
SNR dB8e+01 dB
Dynamic range43.8
Smoothness0.2903
Saturated0.0%
X-outliers92

Integrity & artefacts

NaN ratio0.00%
Inf count0
Zero ratio0.00%
Spike count130,139
Spike rate10.76%
Jump count56,740
Jump rate4.69%
Clip fraction0.00%

Shape & reference

Baseline slope-14.921
Curvature RMS0.28573
D1 RMS0.25817
RMS to mean1.8285
RMS p954.4196
SAM to mean0.036596
SAM p950.078343
Affine offset p952.7678
Affine gain p95 Δ0.1557
Affine residual p951.5556
Xcorr lag p950

Outliers & repeatability

PCA Q p95/median2.3
Hotelling T2 p95/median3.7
Mahalanobis H p95/median1.9
Repeat groups0

Dimensionality (PCA)

Effective rank2.3
PCs → 95% var3
PCs → 99% var9
Top-10 cum. var99.2%
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_ratio0.000165%0.00faibleNormalExport, saturationcount(X == 0) / count(finite X)alert = min(1, zero_ratio / 0.05)
Amplitude globaleMean reflectanceamplitude.mean_reflectance23.5150.68moyenValeur 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_curve511850.68moyenValeur 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_peak43.8470.00faibleVariabilité forteSaturationmax(mean_spectrum) - min(mean_spectrum)alert increases when dynamic range is abnormally flat
Amplitude globaleVarianceamplitude.variance262.760.00faibleNormal ou hétérogèneMauvais contactvar(X finite)alert increases when variance/dynamic range is abnormally flat
BruitNoise RMSnoise.noise_rms0.00256090.00faibleStableLampe, détecteurmedian MAD(second derivative) * 1.4826 / sqrt(6)alert = noise_rms / signal_scale, saturated at 5%
BruitSNRnoise.snr9182.50.00faibleBon signalAcquisitionmean(abs(X)) / noise_rmsalert decreases with SNR dB; >=40 dB is treated as low alert
BruitBandwise SNRnoise.bandwise_snr_min8.07970.48moyenZone problématiqueDétecteurmin(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_count130,1391.00fortArtefactsCosmic rays, splicecount robust outliers in second derivativealert follows spike_rate, saturated at 1%
Artefacts locauxSpike rateartefacts.spike_rate10.8%1.00fortSpectre suspectInterpolationspike_count / (n_samples * (n_features - 2))alert = min(1, spike_rate / 0.01)
Artefacts locauxJump countartefacts.jump_count56,7401.00fortRaccord détecteurSplicecount robust outliers in first derivativealert follows jump_rate, saturated at 1%
Artefacts locauxJump rateartefacts.jump_rate4.69%1.00fortProblème spectralCalibrationjump_count / (n_samples * (n_features - 1))alert = min(1, jump_rate / 0.01)
Artefacts locauxClip fractionartefacts.clip_fraction0.000165%0.00faibleNormalDétecteur saturéfraction of finite cells equal to repeated min/max extremaalert = min(1, clip_fraction / 0.01)
Forme spectraleBaseline slopeshape.baseline_slope-14.9210.68moyenDériveÉclairementlinear slope of mean_spectrum over normalized axisalert = abs(slope / signal_scale), saturated at 0.5
Forme spectraleCurvature RMSshape.curvature_rms0.285730.65moyenForme inhabituelleFond, splicemedian RMS(second derivative per spectrum)alert = curvature_rms / signal_scale, saturated at 1%
Forme spectraleD1 RMSshape.d1_rms0.258170.12faiblePlatBiologie ou artefactmedian RMS(first derivative per spectrum)alert = d1_rms / signal_scale, saturated at 5%
Outliers multivariésPCA Q (SPE)outliers.pca_q_ratio2.34370.29faibleConformeArtefact, mélangep95(Q/SPE residual) / median(Q/SPE residual)alert = min(1, pca_q_ratio / 8)
Outliers multivariésHotelling T²outliers.hotelling_t2_ratio3.70140.46moyenExtrême mais cohérentVariabilité naturellep95(Hotelling T2) / median(Hotelling T2)alert = min(1, hotelling_t2_ratio / 8)
Outliers multivariésMahalanobis Houtliers.mahalanobis_h_ratio1.92390.48moyenOutlier globalDomaine différentp95(sqrt(T2)) / median(sqrt(T2))alert = min(1, mahalanobis_h_ratio / 4)
Comparaison à référenceRMS to mean spectrumreference.rms_to_mean_spectrum_p954.41960.40faibleTypiqueDomain 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_p950.0783430.22faibleSimilaireFond, 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_density0.0493330.84fortSous-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_p952.51070.76fortSpectre 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_p950.563620.84fortSpectre atypiqueDiverses causesp95 IsolationForest anomaly score on PCA scoresalert follows structure complexity; raw score is implementation-dependent
X PCA score plot-400-2000200400600-300-200-1000100200PC1 -87.68 · PC2 -5.271PC1 365.5 · PC2 -55.71PC1 174.5 · PC2 36.49PC1 -49.98 · PC2 63.03PC1 -192 · PC2 9.905PC1 157.5 · PC2 63.62PC1 33.87 · PC2 -62.53PC1 -158.7 · PC2 -16.88PC1 -9.3 · PC2 52.12PC1 59.91 · PC2 31.59PC1 -152.1 · PC2 33.42PC1 70.25 · PC2 40.6PC1 -138 · PC2 -7.082PC1 -130.4 · PC2 85.45PC1 185.9 · PC2 -41.83PC1 -133.2 · PC2 35.69PC1 189.2 · PC2 65.22PC1 14.66 · PC2 -33.34PC1 185.9 · PC2 -34.69PC1 -162 · PC2 50.49PC1 -89.81 · PC2 72.96PC1 -32.44 · PC2 9.845PC1 -64.46 · PC2 71.82PC1 180.4 · PC2 20.05PC1 4.375 · PC2 42.03PC1 9.2 · PC2 -14.76PC1 -127.9 · PC2 28.32PC1 -30.71 · PC2 30.22PC1 -19.49 · PC2 50.15PC1 56.54 · PC2 20.14PC1 101 · PC2 80.88PC1 -67.09 · PC2 -36.49PC1 -45.51 · PC2 -41.95PC1 -137.1 · PC2 50.03PC1 -57.5 · PC2 23.15PC1 -124 · PC2 57.26PC1 21.13 · PC2 -21.23PC1 132.3 · PC2 33.26PC1 -64.16 · PC2 52.99PC1 -7.919 · PC2 19.57PC1 35.79 · PC2 47.3PC1 -21.49 · PC2 76.72PC1 -67.77 · PC2 43.03PC1 -44.88 · PC2 12.23PC1 -23.29 · PC2 -37.69PC1 -98.06 · PC2 30.85PC1 -68.16 · PC2 -6.981PC1 -10.89 · PC2 -70.37PC1 78.24 · PC2 43.46PC1 -55.47 · PC2 88.93PC1 354.5 · PC2 68.58PC1 -78.83 · PC2 20.63PC1 272.1 · PC2 23.23PC1 166.4 · PC2 88.67PC1 4.821 · PC2 -4.659PC1 -49.33 · PC2 61.26PC1 103.2 · PC2 -41.19PC1 -84.64 · PC2 -43.4PC1 -65.18 · PC2 86.11PC1 26.45 · PC2 -26.09PC1 -62.17 · PC2 94.47PC1 105.4 · PC2 -23.09PC1 6.429 · PC2 -10.69PC1 93.88 · PC2 50.22PC1 59.93 · PC2 6.078PC1 72.98 · PC2 -58.9PC1 91.83 · PC2 30.46PC1 413.7 · PC2 88.68PC1 -112.7 · PC2 -73.07PC1 -7.76 · PC2 64.23PC1 54.23 · PC2 36.58PC1 63.5 · PC2 120PC1 -38.52 · PC2 1.623PC1 37.66 · PC2 -15.15PC1 -18.3 · PC2 27.63PC1 57.61 · PC2 71.55PC1 -41.68 · PC2 -8.131PC1 59.67 · PC2 31.82PC1 -145.1 · PC2 57.03PC1 123 · PC2 -5.701PC1 -47.98 · PC2 67.04PC1 -28.77 · PC2 99.27PC1 -106.4 · PC2 -25.47PC1 11.98 · PC2 -72.61PC1 -91.46 · PC2 0.8527PC1 135 · PC2 -5.804PC1 -2.488 · PC2 -7.242PC1 -19.9 · PC2 28.18PC1 137.3 · PC2 50.37PC1 41.08 · PC2 -29.78PC1 258 · PC2 30.87PC1 -124.3 · PC2 -2.608PC1 -130.7 · PC2 46.83PC1 -49.41 · PC2 81.05PC1 -10.58 · PC2 -1.536PC1 -118.8 · PC2 79.39PC1 103.2 · PC2 37.01PC1 10.67 · PC2 108.4PC1 -17.21 · PC2 14.14PC1 -56.39 · PC2 32.57PC1 -97.52 · PC2 11.76PC1 -59.46 · PC2 10.01PC1 -74.28 · PC2 11.47PC1 -20.46 · PC2 -5.041PC1 -21.13 · PC2 -31.88PC1 -51.33 · PC2 41.87PC1 99.66 · PC2 -14.03PC1 -4.786 · PC2 63.6PC1 -169.8 · PC2 9.689PC1 0.1468 · PC2 37PC1 -18.99 · PC2 10.68PC1 51.93 · PC2 54.59PC1 -44.94 · PC2 54.64PC1 64.29 · PC2 80.29PC1 137.4 · PC2 -20.84PC1 111.1 · PC2 62.54PC1 -70.57 · PC2 78.26PC1 -86.14 · PC2 50.6PC1 -74.22 · PC2 97.59PC1 2.564 · PC2 30.75PC1 -2.705 · PC2 31.8PC1 -94.8 · PC2 40.45PC1 -191.6 · PC2 -13.63PC1 -55.75 · PC2 -5.21PC1 -89.85 · PC2 87.58PC1 -62.29 · PC2 39.93PC1 61.67 · PC2 34.07PC1 39.76 · PC2 28.3PC1 70.89 · PC2 18.81PC1 211 · PC2 45.03PC1 -43.49 · PC2 -10.86PC1 36.86 · PC2 68.39PC1 -94.45 · PC2 70.17PC1 122 · PC2 -0.02191PC1 282.7 · PC2 117.2PC1 -46.31 · PC2 68.91PC1 -18.87 · PC2 -29.89PC1 -53.24 · PC2 0.8925PC1 32.5 · PC2 -87.99PC1 -15.54 · PC2 -25.35PC1 37.59 · PC2 -14.88PC1 -55.62 · PC2 44.53PC1 -31.17 · PC2 45.71PC1 -68.08 · PC2 28.52PC1 8.618 · PC2 11.06PC1 -62.35 · PC2 7.384PC1 -10.84 · PC2 54.86PC1 24.49 · PC2 6.927PC1 215.6 · PC2 50.19PC1 20.53 · PC2 21.82PC1 -104.4 · PC2 8.738PC1 39.59 · PC2 52.56PC1 -89.89 · PC2 24.52PC1 -57.47 · PC2 55.84PC1 2.834 · PC2 45.48PC1 -262.9 · PC2 10.43PC1 -155.1 · PC2 5.273PC1 129 · PC2 107.7PC1 -59.66 · PC2 40.33PC1 20.6 · PC2 40.14PC1 -53.13 · PC2 -8.065PC1 -40.11 · PC2 -8.597PC1 111.2 · PC2 71.43PC1 33.38 · PC2 -37.39PC1 36.57 · PC2 98.24PC1 65.55 · PC2 -10.26PC1 -55.83 · PC2 4.216PC1 76.76 · PC2 -76.74PC1 -25.59 · PC2 -70.96PC1 -26.74 · PC2 19.01PC1 -154.6 · PC2 -11.09PC1 44.28 · PC2 -113.5PC1 105.8 · PC2 76.46PC1 15.69 · PC2 40.41PC1 -57.31 · PC2 -35.36PC1 -68.95 · PC2 59.1PC1 17.51 · PC2 93.95PC1 -52.28 · PC2 -18.76PC1 -165.5 · PC2 22.79PC1 -118.5 · PC2 29.77PC1 -133.3 · PC2 4.839PC1 -151.3 · PC2 -69.22PC1 -168.4 · PC2 50.9PC1 -166.2 · PC2 8.665PC1 -58.75 · PC2 -79.99PC1 -216.9 · PC2 -77.7PC1 -28.55 · PC2 43.97PC1 -130.6 · PC2 9.082PC1 -155.7 · PC2 36.26PC1 -187.5 · PC2 -6.725PC1 -246.1 · PC2 -33.74PC1 -29.16 · PC2 36.4PC1 -77.29 · PC2 4.734PC1 -99.13 · PC2 1.296PC1 -201.5 · PC2 -63.48PC1 -216.1 · PC2 28.42PC1 -180.2 · PC2 -44.2PC1 -166.3 · PC2 26.71PC1 -21.88 · PC2 46.44PC1 -65.58 · PC2 58.39PC1 -17.92 · PC2 65.42PC1 -103.6 · PC2 -10.84PC1 -42.42 · PC2 59.38PC1 -109.2 · PC2 20.75PC1 -91.9 · PC2 112.1PC1 -26.65 · PC2 32.75PC1 -54.57 · PC2 33.19PC1 -40.32 · PC2 40.32PC1 -0.1837 · PC2 17.18PC1 -127.6 · PC2 64.95PC1 -136.7 · PC2 64.63PC1 -34.24 · PC2 26.61PC1 94.57 · PC2 69.12PC1 -48.23 · PC2 21.08PC1 33.35 · PC2 19.71PC1 -30.72 · PC2 11.07PC1 -101.2 · PC2 -12.4PC1 -90.2 · PC2 71.55PC1 -64.5 · PC2 63.28PC1 26.45 · PC2 0.7801PC1 -52.6 · PC2 -9.746PC1 133 · PC2 34.47PC1 52.23 · PC2 -0.3345PC1 14.51 · PC2 58.16PC1 -11.91 · PC2 120.8PC1 -74.9 · PC2 34.54PC1 -29 · PC2 44.92PC1 10.24 · PC2 32.37PC1 46.1 · PC2 -8.52PC1 -31.52 · PC2 -1.316PC1 -53.24 · PC2 76.58PC1 31.3 · PC2 37.72PC1 -103.4 · PC2 10.63PC1 -210.3 · PC2 95.59PC1 -104.4 · PC2 16.54PC1 8.119 · PC2 79.68PC1 0.1455 · PC2 99.48PC1 117.7 · PC2 81.92PC1 64.83 · PC2 49.52PC1 18.26 · PC2 -38.88PC1 -138.3 · PC2 41.42PC1 -74.14 · PC2 -37.85PC1 -155.3 · PC2 28.13PC1 -85.16 · PC2 30.19PC1 -34.57 · PC2 77.46PC1 -46.77 · PC2 -21.25PC1 -98.72 · PC2 -18.84PC1 -90.3 · PC2 67.06PC1 -15.42 · PC2 -4.351PC1 -162 · PC2 35.98PC1 -52.46 · PC2 63.26PC1 -28.72 · PC2 83.4PC1 -113.5 · PC2 39.83PC1 -75.82 · PC2 51.99PC1 116.7 · PC2 84PC1 69.76 · PC2 26.48PC1 45.22 · PC2 91.62PC1 -36.61 · PC2 2.922PC1 5.615 · PC2 -29.4PC1 126.9 · PC2 13.05PC1 -67.27 · PC2 -15.98PC1 155 · PC2 -54.03PC1 73.42 · PC2 -40.79PC1 -65.75 · PC2 56.44PC1 -152.2 · PC2 69.39PC1 56.99 · PC2 -23.62PC1 -61.84 · PC2 -79.32PC1 48.65 · PC2 1.585PC1 -98.33 · PC2 -60.47PC1 81.66 · PC2 -28.6PC1 88.51 · PC2 27.96PC1 9.97 · PC2 -8.55PC1 -19.24 · PC2 -12.36PC1 24.88 · PC2 -28.89PC1 119.6 · PC2 -39.78PC1 20.98 · PC2 -5.098PC1 3.479 · PC2 -81.16PC1 -17.81 · PC2 -9.798PC1 283.9 · PC2 25.53PC1 -10.93 · PC2 -24PC1 -83.58 · PC2 20.5PC1 60.24 · PC2 -68.96PC1 7.037 · PC2 26.17PC1 50.09 · PC2 -107.5PC1 -70.74 · PC2 -11.66PC1 -80.3 · PC2 62.72PC1 39.33 · PC2 -10.98PC1 25.67 · PC2 -14.41PC1 156.9 · PC2 -39.44PC1 63.79 · PC2 -17.45PC1 -13.68 · PC2 -48.78PC1 99.85 · PC2 -51.24PC1 147.4 · PC2 57.55PC1 -10.64 · PC2 -32.35PC1 123.8 · PC2 -221.9PC1 60.73 · PC2 1.091PC1 93.17 · PC2 26.07PC1 208.3 · PC2 -81.1PC1 9.466 · PC2 -15.23PC1 84.08 · PC2 -13.08PC1 -4.192 · PC2 -21.15PC1 21.73 · PC2 -71.33PC1 70.75 · PC2 -69.58PC1 -21.73 · PC2 -15.24PC1 300.1 · PC2 -3.914PC1 -54.5 · PC2 -49.37PC1 156.2 · PC2 -12.93PC1 -115.3 · PC2 -9.843PC1 42.24 · PC2 14.74PC1 6.201 · PC2 30.87PC1 98.89 · PC2 42.38PC1 83.09 · PC2 58.39PC1 -111.6 · PC2 -59.78PC1 -71.79 · PC2 -38.59PC1 77.38 · PC2 -98.35PC1 47.31 · PC2 -56.43PC1 12.37 · PC2 60.55PC1 19.71 · PC2 -72.5PC1 42.08 · PC2 -22.87PC1 -42.42 · PC2 -64.24PC1 -92.75 · PC2 -87.28PC1 -235.4 · PC2 -108PC1 -54.18 · PC2 -75.97PC1 88.07 · PC2 28.05PC1 -69.49 · PC2 8.145PC1 63.49 · PC2 56.84PC1 39.36 · PC2 27.07PC1 -21.59 · PC2 -2.255PC1 13.03 · PC2 -32.99PC1 102.7 · PC2 -31.4PC1 61.14 · PC2 32.55PC1 -109.1 · PC2 96.3PC1 23.71 · PC2 -48.54PC1 -24.6 · PC2 10.65PC1 152.6 · PC2 11.51PC1 182.7 · PC2 -41.61PC1 -26.25 · PC2 -23.21PC1 14.46 · PC2 -39.76PC1 26.14 · PC2 -19.7PC1 29.15 · PC2 -27.15PC1 228 · PC2 -1.112PC1 9.123 · PC2 -2.467PC1 206.6 · PC2 -50.79PC1 144.3 · PC2 -39.25PC1 -76.47 · PC2 -15.83PC1 -58.1 · PC2 -56.15PC1 -58.41 · PC2 -55.43PC1 -13.08 · PC2 -84.39PC1 -32.85 · PC2 -50.6PC1 67.48 · PC2 14.93PC1 -99.65 · PC2 -78.25PC1 8.096 · PC2 13.62PC1 30.42 · PC2 -26.22PC1 -74.47 · PC2 -5.941PC1 66.4 · PC2 -105.7PC1 6.883 · PC2 -16.19PC1 -2.395 · PC2 -53.26PC1 59.66 · PC2 -14.52PC1 -111.9 · PC2 -55.44PC1 -11.44 · PC2 60.83PC1 -35.79 · PC2 -28.26PC1 149.5 · PC2 -99.56PC1 53.75 · PC2 47.56PC1 -69.48 · PC2 5.454PC1 13.5 · PC2 -53.33PC1 -6.322 · PC2 -55.95PC1 -51.68 · PC2 4.405PC1 22.12 · PC2 79.71PC1 200.9 · PC2 34.08PC1 9.053 · PC2 6.889PC1 143.4 · PC2 -79.13PC1 102.6 · PC2 -38.84PC1 225.6 · PC2 -102.6PC1 29.5 · PC2 -73.44PC1 -72.91 · PC2 23.12PC1 -4.073 · PC2 36.87PC1 -68.36 · PC2 38.2PC1 41.03 · PC2 -43.66PC1 37.98 · PC2 -12.93PC1 15.82 · PC2 -2.696PC1 -66.49 · PC2 -19.51PC1 -24.83 · PC2 -7.905PC1 99.27 · PC2 -11.88PC1 58.79 · PC2 20.72PC1 -24.78 · PC2 -15.34PC1 -36.1 · PC2 12.07PC1 96.49 · PC2 -57.8PC1 -39.12 · PC2 -88.08PC1 14.4 · PC2 -62.28PC1 149.5 · PC2 -21.54PC1 127 · PC2 -3.983PC1 -96.74 · PC2 49.77PC1 -125.1 · PC2 -55.33PC1 42.47 · PC2 -19.21PC1 34.5 · PC2 -39.11PC1 -2.776 · PC2 -58.71PC1 27.56 · PC2 -81.51PC1 -91.88 · PC2 -21.16PC1 90.38 · PC2 49.62PC1 -2.054 · PC2 -93.91PC1 75.52 · PC2 -27.93PC1 79.53 · PC2 17.75PC1 -6.231 · PC2 -63.1PC1 131 · PC2 0.1988PC1 120.7 · PC2 -149.8PC1 32.56 · PC2 -125.3PC1 -126.1 · PC2 -20.65PC1 37.76 · PC2 -100.1PC1 -87.95 · PC2 -68.63PC1 44.83 · PC2 -108.6PC1 123.7 · PC2 -61.18PC1 13.99 · PC2 -46.68PC1 -22.53 · PC2 -10.45PC1 87.28 · PC2 -48.75PC1 187.6 · PC2 15.71PC1 -4.368 · PC2 -28.86PC1 -19.47 · PC2 -37.07PC1 -53.56 · PC2 25.67PC1 5.112 · PC2 -5.108PC1 61.19 · PC2 -17.23PC1 96.84 · PC2 -63.11PC1 46.68 · PC2 -2.131PC1 52.96 · PC2 -21.33PC1 74.86 · PC2 12.97PC1 -66.46 · PC2 -35.64PC1 231.4 · PC2 -24.74PC1 43.77 · PC2 -13.3PC1 44.96 · PC2 -6.217PC1 48.9 · PC2 -101.9PC1 -46.67 · PC2 -17.47PC1 45.24 · PC2 -19.77PC1 -3.765 · PC2 -23.2PC1 -70.42 · PC2 -86.32PC1 59.26 · PC2 -0.315PC1 2.367 · PC2 26.82PC1 479.9 · PC2 145PC1 43.81 · PC2 -47.7PC1 99.23 · PC2 -53.51PC1 46.43 · PC2 -63.34PC1 142.5 · PC2 -62.54PC1 117 · PC2 -11.32PC1 -104.1 · PC2 -58.23PC1 -67.81 · PC2 -32.36PC1 -161.5 · PC2 -57.6PC1 17.91 · PC2 -71.61PC1 -20.14 · PC2 62.05PC1 1.632 · PC2 -5.876PC1 -40.14 · PC2 -14.78PC1 -127.8 · PC2 -33.21PC1 -42.76 · PC2 -61.41PC1 -116.6 · PC2 21.88PC1 93.93 · PC2 -34.33PC1 -57.44 · PC2 -8.269PC1 -53.71 · PC2 34.68PC1 114 · PC2 -40.55PC1 72.48 · PC2 -19.88PC1 -43.83 · PC2 -14.71PC1 156.1 · PC2 -55.55PC1 -84.83 · PC2 -8.297PC1 161.2 · PC2 25.76PC1 -2.12 · PC2 -35.67PC1 -27.87 · PC2 -60.17PC1 53.18 · PC2 29.99PC1 188.8 · PC2 2.484PC1 24.51 · PC2 -77.04PC1 -70.33 · PC2 79.31PC1 39.04 · PC2 34.9PC1 -34.45 · PC2 -10.02PC1 25.69 · PC2 -119.4PC1 -116.8 · PC2 -46.02PC1 32.97 · PC2 31.69PC1 -130.3 · PC2 1.494PC1 -83.86 · PC2 -60.1PC1 -105.8 · PC2 -33.1PC1 -21.91 · PC2 -50.24PC1 -39.99 · PC2 -5.13PC1 -98.16 · PC2 -86.31PC1 -111 · PC2 -57.43PC1 -49.97 · PC2 -28.75PC1 -41.72 · PC2 8.806PC1 -97.68 · PC2 -71.85PC1 63.5 · PC2 -12PC1 -48.45 · PC2 -18.31PC1 125.9 · PC2 -19.91PC1 110.5 · PC2 63.86PC1 -68.47 · PC2 -74.17PC1 94.74 · PC2 61.75PC1 139.2 · PC2 -56.26PC1 -98.09 · PC2 2.916PC1 0.4006 · PC2 18.05PC1 -58.09 · PC2 -29.2PC1 -41.71 · PC2 43.18PC1 -59.26 · PC2 -20.36PC1 60.21 · PC2 -14.7PC1 -190.2 · PC2 -56.46PC1 28.45 · PC2 68.62PC1 65.16 · PC2 9.866PC1 51.93 · PC2 -62.8PC1 55.07 · PC2 34.52PC1 -36.4 · PC2 -8.338PC1 -34.16 · PC2 -24.64PC1 -3.918 · PC2 -12.85PC1 -29.83 · PC2 -28.3PC1 -51.8 · PC2 50.58PC1 -111 · PC2 -23.29PC1 -27.5 · PC2 -67.83PC1 -75.73 · PC2 -69.33PC1 55.5 · PC2 5.482PC1 55 · PC2 64.31PC1 -14.51 · PC2 21.04PC1 -63.28 · PC2 -24.8PC1 -118.1 · PC2 -2.487PC1 -12.03 · PC2 -31.45PC1 123.5 · PC2 1.679PC1 -114 · PC2 -0.3021PC1 24.48 · PC2 27.06PC1 -6.466 · PC2 -57.4PC1 17.72 · PC2 33.22PC1 162.9 · PC2 162.2PC1 75 · PC2 -30.47PC1 -44.28 · PC2 -72.07PC1 53.55 · PC2 -22.91PC1 -101.6 · PC2 -50.31PC1 -68.43 · PC2 -21.86PC1 -65.55 · PC2 -87.46PC1 9.529 · PC2 -10.98PC1 62.96 · PC2 -23.23PC1 85.56 · PC2 -74.46PC1 81.55 · PC2 11.88PC1 -58.13 · PC2 44.33PC1 -102.6 · PC2 -26.72PC1 39.58 · PC2 44.54PC1 7.854 · PC2 -25.75PC1 -64.83 · PC2 -87.08PC1 308.8 · PC2 -31.45PC1 19.7 · PC2 -32PC1 -121.2 · PC2 -27.57PC1 -75.53 · PC2 16.98PC1 124.3 · PC2 30.68PC1 -90.01 · PC2 -8.016PC1 -130.7 · PC2 22.06PC1 -45.19 · PC2 -13.2PC1 93.97 · PC2 -31.41PC1 53.11 · PC2 -33.37PC1 93.84 · PC2 -59.63PC1 22.03 · PC2 -13.57PC1 20.95 · PC2 -92.55PC1 -91.44 · PC2 -35.82PC1 32.09 · PC2 -66.57PC1 -41.47 · PC2 -89.72PC1 29.36 · PC2 18.11PC1 11.29 · PC2 -66.19PC1 -6.546 · PC2 35.13PC1 -91.69 · PC2 -43.96PC1 0.529 · PC2 -29.6PC1 138 · PC2 -66.79PC1 (73.5%)PC2 (19.4%)556 scores
PCA explained variance0%25%50%75%100%PC1: 73.5% (cumulative 73.5%)1PC2: 19.4% (cumulative 93.0%)2PC3: 3.9% (cumulative 96.9%)3PC4: 0.9% (cumulative 97.8%)4PC5: 0.5% (cumulative 98.3%)5PC6: 0.3% (cumulative 98.6%)6PC7: 0.2% (cumulative 98.8%)7PC8: 0.1% (cumulative 99.0%)8PC9: 0.1% (cumulative 99.1%)9PC10: 0.1% (cumulative 99.2%)10cumulative explained variancePC variancecumulativeprincipal component · cumulative (dashed)
X-Y spectral correlation 4
X · chloride_average spectral correlation-1-0.500.51absolute correlation envelopesigned correlationabsolute correlation01,0002,0003,000|r|signed raxis · Pearson correlation scale
X · chloridometer_reading1 spectral correlation-1-0.500.51absolute correlation envelopesigned correlationabsolute correlation01,0002,0003,000|r|signed raxis · Pearson correlation scale
X · chloridometer_reading2 spectral correlation-1-0.500.51absolute correlation envelopesigned correlationabsolute correlation01,0002,0003,000|r|signed raxis · Pearson correlation scale
Targetmax |r|axis @ maxmean |r||r| ≥ .5
chloride_average0.2667070.08140.0%
chloridometer_reading10.2697060.08130.0%
chloridometer_reading20.2717060.08260.0%
chloridometer_reading30.2717060.0830.0%

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 4

chloride_average

target · numeric
chloride_average distribution01002000 – 331.1: 163331.1 – 662.1: 53662.1 – 993.2: 39993.2 – 1324: 341324 – 1655: 281655 – 1986: 281986 – 2317: 322317 – 2648: 332648 – 2980: 232980 – 3311: 313311 – 3642: 223642 – 3973: 223973 – 4304: 134304 – 4635: 114635 – 4966: 104966 – 5297: 45297 – 5628: 65628 – 5959: 05959 – 6290: 06290 – 6621: 26621 – 6952: 06952 – 7283: 07283 – 7614: 07614 – 7945: 202,0004,0006,0008,000
n / missing556 / 0
Mean ± SD1664 ± 1.53e+03
Median1218
Range0 – 7945
CV0.92
Skew / kurtosis0.92 / 0.29
Normal?no

chloridometer_reading1

target · numeric
chloridometer_reading1 distribution01002000 – 11.71: 16211.71 – 23.42: 5423.42 – 35.12: 4135.12 – 46.83: 3146.83 – 58.54: 3058.54 – 70.25: 2670.25 – 81.96: 2981.96 – 93.67: 3793.67 – 105.4: 25105.4 – 117.1: 30117.1 – 128.8: 19128.8 – 140.5: 24140.5 – 152.2: 12152.2 – 163.9: 11163.9 – 175.6: 9175.6 – 187.3: 4187.3 – 199: 7199 – 210.8: 0210.8 – 222.5: 0222.5 – 234.2: 2234.2 – 245.9: 0245.9 – 257.6: 0257.6 – 269.3: 0269.3 – 281: 20100200300
n / missing556 / 1
Mean ± SD58.84 ± 54.1
Median43
Range0 – 281
CV0.92
Skew / kurtosis0.92 / 0.31
Normal?no

chloridometer_reading2

target · numeric
chloridometer_reading2 distribution01002000 – 11.79: 15811.79 – 23.58: 5923.58 – 35.38: 3835.38 – 47.17: 3347.17 – 58.96: 2858.96 – 70.75: 2970.75 – 82.54: 3482.54 – 94.33: 3094.33 – 106.1: 26106.1 – 117.9: 27117.9 – 129.7: 23129.7 – 141.5: 23141.5 – 153.3: 13153.3 – 165.1: 10165.1 – 176.9: 10176.9 – 188.7: 4188.7 – 200.5: 6200.5 – 212.2: 0212.2 – 224: 1224 – 235.8: 1235.8 – 247.6: 0247.6 – 259.4: 0259.4 – 271.2: 0271.2 – 283: 20100200300
n / missing556 / 1
Mean ± SD59.26 ± 54.5
Median43
Range0 – 283
CV0.919
Skew / kurtosis0.91 / 0.29
Normal?no

chloridometer_reading3

target · numeric
chloridometer_reading3 distribution01002000 – 11.75: 15911.75 – 23.5: 5923.5 – 35.25: 3535.25 – 47: 3547 – 58.75: 2758.75 – 70.5: 2770.5 – 82.25: 3482.25 – 94: 3194 – 105.8: 25105.8 – 117.5: 31117.5 – 129.2: 22129.2 – 141: 21141 – 152.8: 14152.8 – 164.5: 11164.5 – 176.2: 9176.2 – 188: 4188 – 199.8: 7199.8 – 211.5: 0211.5 – 223.2: 0223.2 – 235: 2235 – 246.8: 0246.8 – 258.5: 0258.5 – 270.2: 0270.2 – 282: 20100200300
n / missing556 / 1
Mean ± SD59.21 ± 54.3
Median43
Range0 – 282
CV0.918
Skew / kurtosis0.91 / 0.28
Normal?no

Metadata 6

scan_id

metadata · categorical
n / missing556 / 0
Classes556
Balance (entropy)1
Imbalance ratio1
Top classHR.060623.0000.sig (1)

trial

metadata · categorical
trial classestrial02trial02: 297297trial01trial01: 259259
n / missing556 / 0
Classes2
Balance (entropy)1
Imbalance ratio1
Top classtrial02 (297)

pot_number

metadata · numeric
pot_number distribution01020301 – 14.46: 2714.46 – 27.92: 2527.92 – 41.38: 2841.38 – 54.83: 2554.83 – 68.29: 2768.29 – 81.75: 2481.75 – 95.21: 2795.21 – 108.7: 23108.7 – 122.1: 26122.1 – 135.6: 26135.6 – 149: 27149 – 162.5: 25162.5 – 176: 23176 – 189.4: 27189.4 – 202.9: 25202.9 – 216.3: 26216.3 – 229.8: 23229.8 – 243.2: 25243.2 – 256.7: 25256.7 – 270.2: 24270.2 – 283.6: 10283.6 – 297.1: 12297.1 – 310.5: 13310.5 – 324: 130100200300400
n / missing556 / 0
Mean ± SD147.9 ± 87.7
Median146
Range1 – 324
CV0.593
Skew / kurtosis0.11 / -1.1
Normal?no

genotype

metadata · categorical
genotype classesDOAN9024DOAN9024: 3030LONGII9018LONGII9018: 3030NM11-085NM11-085: 3030NM11-081NM11-081: 3030AZ11-012BAZ11-012B: 3030NM11-072NM11-072: 3030NM11-062NM11-062: 2929BERL9043BERL9043: 2929GIRDIANA6BGIRDIANA6B: 2929NM11-084NM11-084: 2929+10 more+10 more: 233233
n / missing556 / 0
Classes23
Balance (entropy)0.98
Imbalance ratio4
Top classDOAN9024 (30)

replicate

metadata · numeric
replicate distribution0501001501 – 1.167: 1101.167 – 1.333: 01.333 – 1.5: 01.5 – 1.667: 01.667 – 1.833: 01.833 – 2: 02 – 2.167: 1112.167 – 2.333: 02.333 – 2.5: 02.5 – 2.667: 02.667 – 2.833: 02.833 – 3: 03 – 3.167: 1123.167 – 3.333: 03.333 – 3.5: 03.5 – 3.667: 03.667 – 3.833: 03.833 – 4: 04 – 4.167: 1134.167 – 4.333: 04.333 – 4.5: 04.5 – 4.667: 04.667 – 4.833: 04.833 – 5: 11012345
n / missing556 / 0
Mean ± SD3.004 ± 1.41
Median3
Range1 – 5
CV0.47
Skew / kurtosis-0.0064 / -1.3
Normal?no

salt_treatment_mM_NaCl

metadata · numeric
salt_treatment_mM_NaCl distribution01002000 – 3.125: 1813.125 – 6.25: 06.25 – 9.375: 09.375 – 12.5: 012.5 – 15.62: 015.62 – 18.75: 018.75 – 21.88: 021.88 – 25: 025 – 28.12: 028.12 – 31.25: 031.25 – 34.38: 034.38 – 37.5: 037.5 – 40.62: 040.62 – 43.75: 043.75 – 46.88: 046.88 – 50: 050 – 53.12: 18453.12 – 56.25: 056.25 – 59.38: 059.38 – 62.5: 062.5 – 65.62: 065.62 – 68.75: 068.75 – 71.88: 071.88 – 75: 191020406080
n / missing556 / 0
Mean ± SD42.31 ± 31.2
Median50
Range0 – 75
CV0.737
Skew / kurtosis-0.42 / -1.5
Normal?no
Constant metadata 14
  • article_urlhttps://acsess.onlinelibrary.wiley.com/doi/10.1002/ppj2.70012
  • doi_publication10.1002/ppj2.70012
  • instrument_nameSVC HR-1024i
  • spectroscopy_typeVIS-NIR
  • signal_typereflectance
  • axis_unitnm
  • axis_min339
  • axis_max2,515
  • n_points_original1,023
  • n_points_interpolated2,177
  • licenseNOASSERTION
  • rights_statusno_explicit_repository_license
  • public_release_allowedFalse
  • decoderpandas.read_csv, numpy.interp

Alignment

Alignment levelobservation
Sample id availableyes
Samples556
Observations (total)556
Reps per samplemin 1 · mean 1 · max 1

Provenance & citation

ContributorHyperspectral sensing for high-throughput chloride detection in grapevines
Origin · script [manual]source_to_standard.py — standardization script (maintainer-only)

Governance & integrity

Tierprivate
LicenseLicenseRef-not-cleared
Permitted useResearch and benchmarking; private use only.
Access policyManual download / private-use-only per source.
RedistributionWiley article is Open Access and points to GitHub data, but the GitHub repository API reports no explicit repository license.
Content version1.0.0
Schema / protocol2.0
Content hashb4a6f0420e815f79…
Processing hashbd157f63e8e4f6ff…
Metadata hashc042c92a6b39d49b…

Load this dataset

# pip install nirs4all-datasets
from nirs4all_datasets import get

# private dataset — export requires a Dataverse token
ds = get("github_grapevine_chloride_svc_hr1024i_visnir", 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.