metadata.json

{
  "READY": true,
  "name": "Noc_s",
  "task": "regression",
  "cat_idx": [],
  "num_samples": 72,
  "num_features": 380,
  "description": "PTR-ToF-MS analysis of hazelnut paste (Corylus avellana L.) from different geographical origins, roasted at varying times and temperatures (72 samples, 380 peaks, 24 batches of technical replicates). Regression: predicted value is the roasting time. S/P=0.19. Experimental details in Mazzucotelli et al. Features are standard-scaled (StandardScaler). 5-fold cross-validation splits are batch-aware: all samples sharing the same measurement batch (IDs field) are kept together in the same fold to prevent data leakage from technical replicates and batch effects.",
  "source": "On the use of TabPFN on mass spectrometry analysis of volatile organic compounds. Scientific Reports (2025). GitHub: https://github.com/CIFASIS/TabPFN-VOCS",
  "label": "Metabolomics",
  "sub_labels": [
    "mass spectrometry",
    "volatile organic compounds"
  ],
  "force_scale": true,
  "split_strategy": "cv"
}