The scikit-learn-native foundation package for chemometrics 🧪 🤖
This is my agri-tech thesis project which compares different non-destructive spectroscopic methods and machine learning models to check seed viability. Thesis can be accessed upon request via link below.
Edge-AI firmware using an ESP32 and AS7263 NIR spectrometer to stream spectral data via Gemini API for non-destructive watermelon ripeness detection
🔬 Enable calibration transfer in vibrational spectroscopy using self-supervised learning to reduce labeled samples and correct instrument differences.
Private model-release staging for PlumSPECTRA trait-specific NIR ensembles and deployment metadata.
Private review dataset for PlumSPECTRA: analysis-ready spectra, phenotypes, folds, and out-of-fold predictions.
Demonstrar que a fusão de dados espectrais e genômicos pode aumentar significativamente a acurácia da predição fenotípica, contribuindo para decisões mais eficientes em programas de seleção
Portable NIR spectroscopy system built with Raspberry Pi and PLS-DA — captures spectral data from cacao pods and analyzes spectral patterns to classify healthy and Black Pod Rot-affected pods for early, non-invasive detection.
Multispectral NIR classification of 11 plastic types: 9-channel ResNet-18 vs. XGBoost on spectral intensity features
NIR spectroscopy-based moisture prediction using PLS, SVR, ANN, and spectral preprocessing techniques.
Closed-form classical surrogate for the ZZ quantum feature map: the induced metric is I + π²Q, the signless Laplacian of the entanglement graph. Code and data for the paper.
SpectraGryph-inspired desktop spectroscopy viewer & processor (PySide6 + pyqtgraph): FTIR/Raman/UV-Vis/NIR/XRF, baseline/smoothing/chemometrics, peak finding & Gaussian/Lorentzian/Voigt fitting, JCAMP-DX/SPC/OPUS IO.
To associate your repository with the nir-spectroscopy topic, visit your repo's landing page and select "manage topics."