Research, software, data and methods. Inspect the context, compare your options, and take a useful next step.
UCI Machine Learning Repository
Printing-process measurements and defect labels for testing robust industrial classification.
Investigate Which missing-data policy produces the most stable cylinder-banding classifier on a frozen split?
Upstream project repository
Versioned research-data management and provenance workflows built on Git and git-annex.
Investigate Does a bounded local workflow record enough input identity and command provenance to identify a changed derived artifact?
UCI Machine Learning Repository
Drive-system measurements for examining sensorless motor-state classification.
Investigate How does drive diagnosis change when acquisition runs rather than individual rows define the split?
Upstream project repository
Dataset loading and transformation tools useful for preserving example identity, split boundaries and preprocessing provenance in AI experiments.
Investigate Can a tiny authored corpus retain row identity and split membership through mapping, filtering and caching?
Upstream project repository
Deep-learning optimization system for investigating distributed training memory, communication and state partitioning.
Investigate Does a selected optimizer-partitioning configuration reproduce a single-process parameter update on a tiny synthetic model?
Upstream project repository
Serialization of Python objects and interpreter state for research workflows.
Investigate Does restoring a generated stateful object preserve a predefined sequence of deterministic observations?
UCI Machine Learning Repository
Bean morphology descriptors for an agricultural multiclass classification and calibration study.
Investigate Which bean varieties remain confused after train-only scaling and classwise uncertainty reporting?
Upstream project repository
Data and pipeline versioning tools for reproducible computational projects.
Investigate Does changing one declared input invalidate exactly the expected stages of a fixed pipeline graph?
UCI Machine Learning Repository
Protein descriptors and cellular-location labels from a microbial classification benchmark.
Investigate How sensitive is E. coli localization performance to feature scaling and class imbalance?
UCI Machine Learning Repository
Ocean and atmospheric observations for investigating temporal forecasting and missing records.
Investigate Does a persistence baseline remain competitive for an El Nino variable under chronological evaluation?
UCI Machine Learning Repository
Simulated electrical-network parameters and stability outcomes for examining surrogate reliability.
Investigate Can a stability classifier generalize to withheld operating-parameter regions?
UCI Machine Learning Repository
Building-shape simulation outputs for studying energy-demand surrogate models.
Investigate How does energy-load prediction change when entire building configurations are held out?
Recorded license evidence is scoped to each source; it is not a blanket permission or a verification result.