Research, software, data and methods. Inspect the context, compare your options, and take a useful next step.
Upstream project repository
Methods for attributing model predictions to input features using Shapley-based ideas.
Investigate Do selected feature attributions match exhaustive Shapley values under a fully specified background distribution?
Upstream project repository
Time-series learning interfaces for forecasting, classification, transformation, and evaluation.
Investigate Does a rolling-origin forecast experiment preserve temporal separation across all transformations and splits?
UCI Machine Learning Repository
Solar-region attributes and flare counts for studying rare-event prediction.
Investigate How well does a count baseline predict flare magnitude when rare outcomes receive explicit uncertainty intervals?
UCI Machine Learning Repository
Crop symptom descriptors and disease labels for investigating categorical agricultural classification.
Investigate How stable is soybean-disease classification when sparse classes and missing fields are reported separately?
UCI Machine Learning Repository
Satellite spectral features and land-cover labels for auditing spatial classification benchmarks.
Investigate Does spatial grouping materially alter Landsat classification error and classwise recall?
UCI Machine Learning Repository
Spacecraft-related measurements for examining extreme class imbalance in fault classification.
Investigate How do minority-class recall and calibration differ from aggregate shuttle classification accuracy?
UCI Machine Learning Repository
Vehicle-shape descriptors for studying invariance in compact image-derived feature spaces.
Investigate How much does feature normalization affect vehicle-silhouette classification on a frozen split?
Upstream project repository
Statistical forecasting models and utilities for collections of time series.
Investigate How do fixed classical forecasting methods compare on held-out synthetic seasonal regimes?
UCI Machine Learning Repository
Industrial electricity measurements for auditing calendar-aware energy forecasts.
Investigate Does a simple seasonal model match a richer steel-industry demand model on the final time block?
UCI Machine Learning Repository
Steel-surface fault descriptors for inspecting industrial multiclass prediction.
Investigate Which steel-fault classes remain difficult after train-only scaling and class-balanced reporting?
UCI Machine Learning Repository
Stone-artifact measurements for investigating small-sample archaeological classification.
Investigate How stable are stone-flake group distinctions under leave-one-artifact-out evaluation?
UCI Machine Learning Repository
Material-composition descriptors and transition temperatures for investigating materials prediction.
Investigate How does superconducting-temperature error change under composition-grouped splitting?
Recorded license evidence is scoped to each source; it is not a blanket permission or a verification result.