← Back to the catalog
Private

FLOPP FTIR polymer classification

plastic · MIR

FLOPP FTIR polymer classification. v2.0 standardized NIRS package: 1 spectral source(s), 1 declared target(s). Auto-generated from dataset_card.json (verify before publication).

nirv2plastic
🔒
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.
186
samples
1,869
wavelengths
1
sources
1
targets
9
metadata
MIR
family

Dataset property explorer

Mean profile risk0.67
Highest axisIntégrité · 1.00
Diagnostics8
Sources profiled1
FLOPP FTIR polymer classification property profile0.250.50.751integritynoiseartefactsbaselinePCA outliersreferencerepeatabilitystructureFLOPP FTIR polymer classification profileintegrity: 1.00noise: 0.00artefacts: 1.00baseline: 1.00PCA outliers: 0.66reference: 0.66repeatability: 0.00structure: 1.00FLOPP FTIR poly…0 center · 1 outer ring · outward = stronger anomaly / heterogeneity signal

Profile axes

Intégrité1.00
Artefacts locaux1.00
Bruit0.00
Outliers PCA0.66
Distance à la référence0.66
Répétabilité0.00
Baseline / forme1.00
Structure multi-régimes1.00
Diagnostic hypotheses00.250.50.751hypothesis scoreSpectre saturé / clippingSpectre saturé / clipping: 0.850.85Splice / raccord détecteursSplice / raccord détecteurs: 0.810.81Erreur calibration / référenc…Erreur calibration / référence blanche: 0.700.70Erreur interpolation / réécha…Erreur interpolation / rééchantillonnage: 0.680.68Fond différentFond différent: 0.640.64Signature VERA25-likeSignature VERA25-like: 0.620.62Différence de sonde / géométr…Différence de sonde / géométrie: 0.550.55Dataset multi-régimesDataset multi-régimes: 0.520.52
DiagnosticScoreForceSignauxInterprétation probable
Spectre saturé / clippingX0.85forteClip fraction 1.00, Baseline/mean/area 1.00, Jump rate 1.00Détecteur saturé ou dynamique insuffisante.
Splice / raccord détecteursX0.81forteSpike rate 1.00, Jump rate 1.00, SNR non dégradé 1.00Rupture aux jonctions de détecteurs, calibration locale ou sonde différente.
Erreur calibration / référence blancheX0.70moyenneBaseline/mean/area 1.00, artefacts locaux 1.00, PCA Q 0.66Décalage systématique entre campagnes, instruments ou référence blanche.
Erreur interpolation / rééchantillonnageX0.68moyenneSpike rate 1.00, Jump rate 1.00, SNR normal/élevé 1.00Artefacts numériques ou traitement spectral incorrect.
Fond différentX0.64moyenneBaseline/mean/area 1.00, PCA Q 0.66, RMS/SAM référence 0.66Effet systématique du support, blanc/noir, transflectance ou environnement de mesure.
Signature VERA25-likeX0.62moyenneSpike rate 1.00, Jump rate 1.00, PCA Q 0.66Combinaison possible changement de sonde + splice, amplifiée par géométrie, fond ou calibration.
Différence de sonde / géométrieX0.55moyenneBaseline/mean/area 1.00, PCA Q 0.66, RMS/SAM référence 0.66Modification de l'illumination, collecte, angle ou distance sonde-échantillon.
Dataset multi-régimesX0.52moyenneStructure PCA 1.00, PCA Q 0.66, RMS/SAM référence 0.66Mélange de campagnes, opérateurs, lots, setups ou sous-populations spectrales.

Spectral sources

FLOPP FTIR spectra

X · MIR · ATR-FTIR instruments as represented in source library
FLOPP FTIR spectra spectra05010015001,0002,0003,0004,0005,000q05-q95 envelopeq25-q75 envelopemedian spectrummedianq25–q75q05–q95wavelength / none399.21none — median 0 (q25–q75 0–0)424.28none — median 0 (q25–q75 0–0)451.28none — median 0 (q25–q75 0–0)476.35none — median 0 (q25–q75 0–0)503.35none — median 0 (q25–q75 0–0)528.42none — median 0 (q25–q75 0–0)555.42none — median 0 (q25–q75 0–0)580.49none — median 0 (q25–q75 0–0)607.49none — median 0 (q25–q75 0–0)632.56none — median 0 (q25–q75 0–0)657.63none — median 0 (q25–q75 0–0)684.63none — median 89.31 (q25–q75 85.52–95.33)709.71none — median 89.16 (q25–q75 83.57–93.51)736.71none — median 90.4 (q25–q75 87.02–95.44)761.78none — median 93.4 (q25–q75 87.05–96.15)788.78none — median 94.67 (q25–q75 91.45–96.63)813.85none — median 95.32 (q25–q75 91.98–96.69)838.92none — median 94.76 (q25–q75 91.57–96.2)865.92none — median 94.82 (q25–q75 90.36–97.2)890.99none — median 95.26 (q25–q75 92.02–97.48)917.99none — median 94.79 (q25–q75 92.43–97.55)943.06none — median 94.84 (q25–q75 91.62–97.66)970.06none — median 93.1 (q25–q75 88.34–96.23)995.13none — median 94.33 (q25–q75 88.6–96.63)1022.1none — median 92.34 (q25–q75 81.37–96.42)1047.2none — median 93.08 (q25–q75 83.26–96.53)1072.3none — median 92.63 (q25–q75 77.75–96.09)1099.3none — median 92.91 (q25–q75 72.15–96.97)1124.3none — median 93.8 (q25–q75 79.25–97.45)1151.3none — median 94 (q25–q75 88.52–96.83)1176.4none — median 93.56 (q25–q75 88.87–96.94)1203.4none — median 93.5 (q25–q75 87.56–96.81)1228.5none — median 91.36 (q25–q75 82.6–96.94)1253.6none — median 91.73 (q25–q75 82.3–96.7)1280.6none — median 93.58 (q25–q75 88–96.82)1305.6none — median 94.41 (q25–q75 90.54–97.37)1332.6none — median 95.03 (q25–q75 91.85–97.55)1357.7none — median 94.67 (q25–q75 91.54–96.65)1384.7none — median 94.75 (q25–q75 92.03–96.87)1409.8none — median 95.16 (q25–q75 88.87–97.53)1436.8none — median 93.8 (q25–q75 91.26–95.55)1461.8none — median 91.47 (q25–q75 87.45–95.24)1486.9none — median 97.03 (q25–q75 93.86–98.73)1513.9none — median 97.98 (q25–q75 95.57–99.26)1539none — median 98.03 (q25–q75 96.14–99.42)1566none — median 98.04 (q25–q75 96.76–99.36)1591.1none — median 97.71 (q25–q75 96.01–99.27)1618.1none — median 97.75 (q25–q75 96.01–99.21)1643.1none — median 97.64 (q25–q75 96–99.05)1670.1none — median 97.57 (q25–q75 94.75–99.08)1695.2none — median 97.83 (q25–q75 95.55–99.44)1720.3none — median 96.4 (q25–q75 89.25–98.82)1747.3none — median 97.44 (q25–q75 93.63–98.99)1772.3none — median 99.1 (q25–q75 97.93–100.2)1799.3none — median 99.3 (q25–q75 98.1–100.2)1824.4none — median 99.48 (q25–q75 98.61–100.4)1851.4none — median 99.67 (q25–q75 98.88–100.5)1876.5none — median 99.71 (q25–q75 98.95–100.5)1901.5none — median 99.78 (q25–q75 98.94–100.7)1928.5none — median 99.86 (q25–q75 98.98–100.7)1953.6none — median 99.82 (q25–q75 98.85–100.7)1980.6none — median 99.84 (q25–q75 98.96–100.7)2005.7none — median 99.87 (q25–q75 99.04–100.7)2032.7none — median 99.92 (q25–q75 99–100.7)2057.8none — median 99.87 (q25–q75 99.02–100.7)2084.8none — median 99.91 (q25–q75 99.03–100.8)2109.8none — median 99.96 (q25–q75 99.04–100.9)2134.9none — median 100 (q25–q75 99.06–100.9)2161.9none — median 100.1 (q25–q75 99.12–101)2187none — median 100 (q25–q75 99.08–101)2214none — median 100 (q25–q75 99.03–101)2239none — median 99.79 (q25–q75 98.44–100.9)2266none — median 99.99 (q25–q75 98.97–100.9)2291.1none — median 100 (q25–q75 99–100.9)2316.2none — median 100.3 (q25–q75 99.18–101.2)2343.2none — median 100.8 (q25–q75 99.65–101.8)2368.3none — median 101 (q25–q75 99.82–102.1)2395.3none — median 100.1 (q25–q75 99.1–101.1)2420.3none — median 100.2 (q25–q75 99.07–101.1)2447.3none — median 100.1 (q25–q75 99.07–101.1)2472.4none — median 100.2 (q25–q75 99.01–101.1)2499.4none — median 100.1 (q25–q75 98.95–101.1)2524.5none — median 100.1 (q25–q75 98.95–101)2549.5none — median 100.1 (q25–q75 98.9–101)2576.5none — median 100.1 (q25–q75 98.88–101)2601.6none — median 100.1 (q25–q75 98.87–101)2628.6none — median 100.1 (q25–q75 98.81–100.9)2653.7none — median 100 (q25–q75 98.81–100.9)2680.7none — median 99.87 (q25–q75 98.64–100.8)2705.8none — median 99.91 (q25–q75 98.63–100.8)2730.8none — median 99.81 (q25–q75 98.55–100.7)2757.8none — median 99.79 (q25–q75 98.51–100.6)2782.9none — median 99.57 (q25–q75 98.42–100.4)2809.9none — median 99.14 (q25–q75 97.76–99.89)2835none — median 97.54 (q25–q75 94.28–99.49)2862none — median 95.59 (q25–q75 91.95–98.13)2887none — median 96.14 (q25–q75 91.59–98.59)2914none — median 93.27 (q25–q75 80.66–96.63)2939.1none — median 94.05 (q25–q75 91.78–96.92)2964.2none — median 97.09 (q25–q75 94.02–98.41)2991.2none — median 98.73 (q25–q75 97.81–99.91)3016.2none — median 99.16 (q25–q75 97.99–100.1)3043.2none — median 99.33 (q25–q75 98.03–100.4)3068.3none — median 99.36 (q25–q75 97.87–100.4)3095.3none — median 99.56 (q25–q75 98.37–100.7)3120.4none — median 99.4 (q25–q75 98.36–100.4)3147.4none — median 99.6 (q25–q75 98.54–100.8)3172.5none — median 99.66 (q25–q75 98.55–100.8)3197.5none — median 99.54 (q25–q75 98.43–100.8)3224.5none — median 99.54 (q25–q75 98.31–100.8)3249.6none — median 99.32 (q25–q75 97.72–100.6)3276.6none — median 99.19 (q25–q75 97.3–100.5)3301.7none — median 99.19 (q25–q75 97.05–100.5)3328.7none — median 99.38 (q25–q75 97.3–100.6)3353.7none — median 99.44 (q25–q75 97.74–100.7)3378.8none — median 99.52 (q25–q75 98.13–100.7)3405.8none — median 99.62 (q25–q75 98.34–100.8)3430.9none — median 99.44 (q25–q75 98.29–100.7)3457.9none — median 99.71 (q25–q75 98.47–100.9)3483none — median 99.73 (q25–q75 98.56–101)3510none — median 99.86 (q25–q75 98.69–101.2)3535none — median 99.83 (q25–q75 98.66–101.1)3562none — median 99.91 (q25–q75 98.8–101.2)3587.1none — median 99.92 (q25–q75 98.9–101.3)3612.2none — median 100.2 (q25–q75 99–101.5)3639.2none — median 100.2 (q25–q75 98.96–101.5)3664.2none — median 100.2 (q25–q75 98.93–101.4)3691.2none — median 100 (q25–q75 98.87–101.3)3716.3none — median 100.2 (q25–q75 99.18–101.5)3743.3none — median 100.4 (q25–q75 99.31–101.8)3768.4none — median 100.3 (q25–q75 99.11–101.6)3793.5none — median 100.3 (q25–q75 99.1–101.5)3820.5none — median 100.3 (q25–q75 99.16–101.6)3845.5none — median 99.97 (q25–q75 98.86–101.2)3872.5none — median 100.1 (q25–q75 98.88–101.3)3897.6none — median 100.1 (q25–q75 98.96–101.4)3924.6none — median 100 (q25–q75 98.92–101.3)3949.7none — median 99.98 (q25–q75 98.86–101.2)3976.7none — median 99.98 (q25–q75 98.87–101.2)4001.7none — median 0 (q25–q75 0–0)

Sampling

Wavelengths1,869
Axis range399.2–4002 none
Mean spacing1.93 none
Griduniform
Observations186

Signal & quality

Value range0 – 114
Mean range0 – 101
Mean level89.16
Area3.214e+05
PTP101.4
Noise RMS0.016289
SNR5.5e+03
SNR dB7e+01 dB
Dynamic range101
Smoothness3.743
Saturated0.0%
X-outliers95

Integrity & artefacts

NaN ratio0.00%
Inf count0
Zero ratio7.70%
Spike count13,177
Spike rate3.79%
Jump count28,911
Jump rate8.32%
Clip fraction7.70%

Shape & reference

Baseline slope49.579
Curvature RMS3.7782
D1 RMS3.1875
RMS to mean4.5682
RMS p958.8117
SAM to mean0.043713
SAM p950.089861
Affine offset p952.5236
Affine gain p95 Δ0.041761
Affine residual p958.0617
Xcorr lag p950

Outliers & repeatability

PCA Q p95/median5.3
Hotelling T2 p95/median3.9
Mahalanobis H p95/median2
Repeat groups0

Dimensionality (PCA)

Effective rank5.6
PCs → 95% var9
PCs → 99% var17
Top-10 cum. var96.7%
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_ratio7.7%1.00fortSpectre tronquéExport, saturationcount(X == 0) / count(finite X)alert = min(1, zero_ratio / 0.05)
Amplitude globaleMean reflectanceamplitude.mean_reflectance89.1631.00fortValeur 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_curve3.2139e+051.00fortValeur 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_peak101.420.00faibleVariabilité forteSaturationmax(mean_spectrum) - min(mean_spectrum)alert increases when dynamic range is abnormally flat
Amplitude globaleVarianceamplitude.variance722.130.00faibleNormal ou hétérogèneMauvais contactvar(X finite)alert increases when variance/dynamic range is abnormally flat
BruitNoise RMSnoise.noise_rms0.0162890.00faibleStableLampe, détecteurmedian MAD(second derivative) * 1.4826 / sqrt(6)alert = noise_rms / signal_scale, saturated at 5%
BruitSNRnoise.snr5473.70.00faibleBon signalAcquisitionmean(abs(X)) / noise_rmsalert decreases with SNR dB; >=40 dB is treated as low alert
BruitBandwise SNRnoise.bandwise_snr_min0—non calculablePas assez d'information pour scorer cette métrique sur ce dataset.Détecteurmin(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_count13,1771.00fortArtefactsCosmic rays, splicecount robust outliers in second derivativealert follows spike_rate, saturated at 1%
Artefacts locauxSpike rateartefacts.spike_rate3.79%1.00fortSpectre suspectInterpolationspike_count / (n_samples * (n_features - 2))alert = min(1, spike_rate / 0.01)
Artefacts locauxJump countartefacts.jump_count28,9111.00fortRaccord détecteurSplicecount robust outliers in first derivativealert follows jump_rate, saturated at 1%
Artefacts locauxJump rateartefacts.jump_rate8.32%1.00fortProblème spectralCalibrationjump_count / (n_samples * (n_features - 1))alert = min(1, jump_rate / 0.01)
Artefacts locauxClip fractionartefacts.clip_fraction7.7%1.00fortClippingDétecteur saturéfraction of finite cells equal to repeated min/max extremaalert = min(1, clip_fraction / 0.01)
Forme spectraleBaseline slopeshape.baseline_slope49.5790.98fortDériveÉclairementlinear slope of mean_spectrum over normalized axisalert = abs(slope / signal_scale), saturated at 0.5
Forme spectraleCurvature RMSshape.curvature_rms3.77821.00fortForme inhabituelleFond, splicemedian RMS(second derivative per spectrum)alert = curvature_rms / signal_scale, saturated at 1%
Forme spectraleD1 RMSshape.d1_rms3.18750.63moyenSpectre structuréBiologie ou artefactmedian RMS(first derivative per spectrum)alert = d1_rms / signal_scale, saturated at 5%
Outliers multivariésPCA Q (SPE)outliers.pca_q_ratio5.31780.66moyenSpectre atypiqueArtefact, mélangep95(Q/SPE residual) / median(Q/SPE residual)alert = min(1, pca_q_ratio / 8)
Outliers multivariésHotelling T²outliers.hotelling_t2_ratio3.92940.49moyenExtrême mais cohérentVariabilité naturellep95(Hotelling T2) / median(Hotelling T2)alert = min(1, hotelling_t2_ratio / 8)
Outliers multivariésMahalanobis Houtliers.mahalanobis_h_ratio1.98230.50moyenOutlier globalDomaine différentp95(sqrt(T2)) / median(sqrt(T2))alert = min(1, mahalanobis_h_ratio / 4)
Comparaison à référenceRMS to mean spectrumreference.rms_to_mean_spectrum_p958.81170.35faibleTypiqueDomain 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_p950.0898610.26faibleSimilaireFond, 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_density0.014371.00fortSous-populationsLots différents1 / median kNN distance in PCA score spacealert follows density_cv/profile structure complexity, not raw density alone
Structure du datasetLocal Outlier Factor (LOF)structure.local_outlier_factor_p952.26880.63moyenSpectre 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_p950.540851.00fortSpectre atypiqueDiverses causesp95 IsolationForest anomaly score on PCA scoresalert follows structure complexity; raw score is implementation-dependent
X PCA score plot-1,00001,0002,000-1,000-5000500PC1 -99.21 · PC2 -48.6PC1 -27.26 · PC2 -46.87PC1 -45.42 · PC2 -63PC1 -33.84 · PC2 -81.29PC1 -37.38 · PC2 -85.07PC1 -97.93 · PC2 -37.84PC1 -58.08 · PC2 -72.52PC1 -143.1 · PC2 -39.48PC1 -43.79 · PC2 -49.76PC1 -124.7 · PC2 -42.56PC1 -132.2 · PC2 -27.72PC1 -92.93 · PC2 -68.61PC1 -161.3 · PC2 -56.08PC1 -79.66 · PC2 -60.58PC1 -36.95 · PC2 -41.11PC1 -125.5 · PC2 -47.7PC1 -99.3 · PC2 -35.67PC1 -110 · PC2 -57.24PC1 -129.9 · PC2 -48.95PC1 -56.7 · PC2 -53.07PC1 70.56 · PC2 -48.17PC1 132.2 · PC2 -73.37PC1 169.1 · PC2 -87.01PC1 39.89 · PC2 -15.8PC1 49.52 · PC2 -31.22PC1 109.3 · PC2 -26.29PC1 -68.37 · PC2 -25.95PC1 77.8 · PC2 -33.58PC1 124.7 · PC2 -57.2PC1 104.8 · PC2 -10.79PC1 -90.67 · PC2 -18.91PC1 -155.8 · PC2 -58.05PC1 -81.14 · PC2 -34.76PC1 6.849 · PC2 -6.937PC1 8.205 · PC2 -62.03PC1 -60.62 · PC2 -50.36PC1 -155.9 · PC2 49.35PC1 -103.1 · PC2 20.1PC1 25.09 · PC2 -73.17PC1 -38.81 · PC2 -55.71PC1 10.59 · PC2 -109.7PC1 21.11 · PC2 -154.2PC1 -30.27 · PC2 -80.18PC1 -127.4 · PC2 13.97PC1 -158.6 · PC2 90.88PC1 -117.4 · PC2 47.24PC1 147 · PC2 90.44PC1 213.5 · PC2 48.85PC1 100.4 · PC2 109.1PC1 -156.8 · PC2 -47.68PC1 124.2 · PC2 80.65PC1 209.3 · PC2 22.82PC1 59.72 · PC2 16.03PC1 213.1 · PC2 51.75PC1 -155.2 · PC2 -73.49PC1 -198.4 · PC2 -60.77PC1 -93.34 · PC2 -128.2PC1 -188 · PC2 -58.24PC1 -165.2 · PC2 -86.72PC1 -66.02 · PC2 -67.71PC1 -191.7 · PC2 -57PC1 -174.5 · PC2 -64.36PC1 -173.6 · PC2 -82.66PC1 -168.7 · PC2 -86.76PC1 -174.3 · PC2 -87.35PC1 -155.6 · PC2 -88.3PC1 -179.2 · PC2 -71.8PC1 -187.1 · PC2 -76.35PC1 -168.3 · PC2 -77.48PC1 -165 · PC2 -92.29PC1 -75.52 · PC2 -124.3PC1 -196 · PC2 -74.82PC1 -48.42 · PC2 9.142PC1 187.1 · PC2 282.3PC1 132.8 · PC2 313.8PC1 164 · PC2 274PC1 23.11 · PC2 153.9PC1 47.5 · PC2 14.09PC1 -72.71 · PC2 36.73PC1 176.1 · PC2 297.4PC1 181.6 · PC2 306PC1 184.7 · PC2 231.8PC1 242.6 · PC2 264.1PC1 83.79 · PC2 176.8PC1 -65.78 · PC2 -84.28PC1 -103.1 · PC2 -37.98PC1 -144.9 · PC2 -41.54PC1 -176.7 · PC2 -10.2PC1 -39.23 · PC2 -82.51PC1 -18.48 · PC2 125.3PC1 -54.6 · PC2 -60.68PC1 -153.1 · PC2 10.21PC1 -83.44 · PC2 -32.92PC1 -93.6 · PC2 -28.56PC1 74.09 · PC2 118.6PC1 73.66 · PC2 161PC1 64.42 · PC2 132.6PC1 82.61 · PC2 167.1PC1 83.44 · PC2 157PC1 98.81 · PC2 126.3PC1 124.3 · PC2 145.6PC1 141.9 · PC2 168.6PC1 -5.356 · PC2 459PC1 223.8 · PC2 186.7PC1 185.8 · PC2 227.5PC1 -107.1 · PC2 34.1PC1 93.7 · PC2 161.7PC1 82.19 · PC2 142.8PC1 225.5 · PC2 272.4PC1 208 · PC2 200.1PC1 167.8 · PC2 303.1PC1 16.6 · PC2 128.9PC1 -9.334 · PC2 260.9PC1 -11.69 · PC2 175.4PC1 28.97 · PC2 -18.84PC1 193.6 · PC2 265.9PC1 -146.9 · PC2 -88.84PC1 -176 · PC2 -58.84PC1 -125.3 · PC2 -100.9PC1 -152.3 · PC2 -95.92PC1 -163.5 · PC2 -67PC1 -185.2 · PC2 -57.03PC1 -180.7 · PC2 -70.59PC1 -162.8 · PC2 -71.69PC1 -126 · PC2 -107.2PC1 -172.9 · PC2 -73.03PC1 -160 · PC2 -74.41PC1 -108.1 · PC2 -79.3PC1 -67.09 · PC2 -81.18PC1 -58.86 · PC2 -99.13PC1 -136.5 · PC2 -105.6PC1 -156.2 · PC2 -74.26PC1 -107.1 · PC2 -116.5PC1 -133.9 · PC2 -103.6PC1 -129.6 · PC2 -64.09PC1 -65.9 · PC2 -78.96PC1 -95.6 · PC2 -100.8PC1 -76.76 · PC2 -89.42PC1 59.22 · PC2 -116.9PC1 71.6 · PC2 -113.7PC1 -123.5 · PC2 -67.66PC1 -36.89 · PC2 -106.6PC1 11.15 · PC2 -114.9PC1 -63.11 · PC2 -93.45PC1 -60.21 · PC2 -103.2PC1 -86.5 · PC2 -68.39PC1 -107.5 · PC2 -69.71PC1 -61.01 · PC2 -102.9PC1 -54.03 · PC2 -80.09PC1 -42.98 · PC2 -2.479PC1 111.9 · PC2 96.25PC1 88.66 · PC2 23.04PC1 564.3 · PC2 -71.39PC1 159.1 · PC2 97.43PC1 258.9 · PC2 30.69PC1 105.8 · PC2 83.64PC1 195.8 · PC2 44.96PC1 179.1 · PC2 103.5PC1 177.7 · PC2 84.66PC1 457.9 · PC2 -49.42PC1 170.6 · PC2 108.2PC1 123 · PC2 27.98PC1 -60.09 · PC2 9.964PC1 -125 · PC2 1.19PC1 -32.99 · PC2 4.159PC1 -31.55 · PC2 58.73PC1 -117.8 · PC2 42.62PC1 -66.61 · PC2 61.19PC1 -82.84 · PC2 -1.718PC1 -106.5 · PC2 6.071PC1 -122.4 · PC2 3.611PC1 24.89 · PC2 52.43PC1 37.63 · PC2 -120.8PC1 117.8 · PC2 43.45PC1 7.883 · PC2 33.07PC1 142.8 · PC2 11.62PC1 105.3 · PC2 40.33PC1 1932 · PC2 -752.1PC1 -110.7 · PC2 -104.7PC1 -25.24 · PC2 -92.48PC1 207.4 · PC2 35.27PC1 398.7 · PC2 -44.45PC1 226.9 · PC2 -18.97PC1 -91.95 · PC2 -49.49PC1 60.05 · PC2 72.86PC1 34.78 · PC2 -24.51PC1 (49.6%)PC2 (19.5%)186 scores
PCA explained variance0%25%50%75%100%PC1: 49.6% (cumulative 49.6%)1PC2: 19.5% (cumulative 69.1%)2PC3: 7.2% (cumulative 76.2%)3PC4: 6.4% (cumulative 82.7%)4PC5: 4.9% (cumulative 87.6%)5PC6: 2.8% (cumulative 90.4%)6PC7: 2.4% (cumulative 92.8%)7PC8: 1.7% (cumulative 94.5%)8PC9: 1.3% (cumulative 95.8%)9PC10: 0.9% (cumulative 96.7%)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

polymer_name

target · categorical
polymer_name classesPPPP: 1919PEPE: 1818RR: 1515ABSABS: 1414CottonCotton: 1414PolyesterPolyester: 1414PSPS: 1414PUPU: 1313NylonNylon: 1212PETPET: 1111+7 more+7 more: 4242
n / missing186 / 0
Classes17
Balance (entropy)0.95
Imbalance ratio19
Top classPP (19)

Metadata 1

sample_description

metadata · categorical
sample_description classesWhite FiberWhite Fiber: 66Pink FiberPink Fiber: 44Yellow FiberYellow Fiber: 44Red FiberRed Fiber: 33Orange FiberOrange Fiber: 22Grey FiberGrey Fiber: 22Purple FiberPurple Fiber: 22Green FiberGreen Fiber: 22Blue FiberBlue Fiber: 22Clear Wrapping FilmClear Wrapping Film: 22+10 more+10 more: 1414
n / missing186 / 1
Classes162
Balance (entropy)0.98
Imbalance ratio6
Top classWhite Fiber (6)
Constant metadata 8
  • library_nameFLOPP
  • spectroscopy_typeFTIR
  • sample_typeplastic particle
  • axis_unitcm^-1
  • axis_min399.2
  • axis_max4002
  • n_points_original1,869
  • signal_typeATR-FTIR intensity

Alignment

Alignment levelsample
Sample id availableyes
Samples186
Observations (total)186
Reps per samplemin 1 · mean 1 · max 1

Splits

originalnot_provided: 186 documented · not applied

Provenance & citation

ContributorATR-FTIR Spectral Libraries of Plastic Particles, FLOPP and FLOPP-e
Origin · url [open]https://acs.figshare.com/articles/dataset/_ATR-FTIR_Spectral_Libraries_of_Plastic_Particles_FLOPP_and_FLOPP-e_for_the_Analysis_of_Microplastics/17070059
Origin · url [open]https://api.figshare.com/v2/articles/17070059
Origin · url [open]https://figshare.com/ndownloader/articles/17070059/versions/1
Origin · figshare [open]10.6084/m9.figshare.17070059 — figshare
Origin · script [manual]source_to_standard.py — standardization script (maintainer-only)
Publication10.1021/acs.analchem.1c02549

Governance & integrity

Tierprivate
LicenseLicenseRef-not-cleared
Permitted useResearch and benchmarking; private use only.
Access policyManual download / private-use-only per source.
RedistributionRights are not cleared for public redistribution, internal/private use only by default.
Content version1.0.0
Schema / protocol2.0
Content hashc861462a52e9d92b…
Processing hash4b07d22cb966349d…
Metadata hash1706d1b2e71e01e6…

Load this dataset

# pip install nirs4all-datasets
from nirs4all_datasets import get

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