Research, software, data and methods. Inspect the context, compare your options, and take a useful next step.
Upstream project repository
Gradient transformation and optimization tools for testing optimizer state and update equations.
Investigate Does one fixed optimizer produce the hand-derived first two updates for a scalar quadratic?
UCI Machine Learning Repository
Network configuration and performance observations for examining engineering response surfaces.
Investigate How does optical-network performance prediction change under configuration-group holdout?
UCI Machine Learning Repository
Atmospheric predictors and ozone-event labels for inspecting temporally dependent rare-event detection.
Investigate How does ozone-event recall change under chronological splitting and train-only imputation?
Upstream project repository
Parameterization and execution of notebooks for repeatable computational workflows.
Investigate Does an isolated parameterized notebook run preserve supplied parameters and produce the independently expected deterministic output?
Upstream project repository
Parameter-efficient model adaptation methods for testing trainable-parameter budgets and adapter composition.
Investigate Does a low-rank adapter preserve the frozen base parameters and produce the same outputs before and after an explicitly supported merge?
UCI Machine Learning Repository
Sensor responses to fragrance samples for examining chemical recognition and repeat-measurement leakage.
Investigate How does perfume recognition change when repeated measurements of a sample stay within one fold?
UCI Machine Learning Repository
Hourly generation measurements for evaluating electricity-supply forecasting across time.
Investigate Does seasonal persistence match a covariate-based generation forecast on a final chronological holdout?
UCI Machine Learning Repository
Hardware challenge-response records for studying generalization in physical-device models.
Investigate How does a challenge-response model generalize when devices or challenge families are held out?
UCI Machine Learning Repository
Generated card hands with rule-based labels for testing rare combinatorial classes.
Investigate Can a poker-hand classifier identify rare hand classes on a split that excludes reordered duplicates?
Upstream project repository
Management and retrieval of data files with registry and checksum support.
Investigate Does a fixed registry detect a deliberately changed input rather than silently accepting stale bytes?
UCI Machine Learning Repository
City electricity-demand measurements for studying seasonal and multi-zone forecasts.
Investigate Does a zone-aware load forecast improve on seasonal persistence in a chronological Tetouan holdout?
UCI Machine Learning Repository
Artificial oscillatory sequences for testing methods against controlled temporal structure.
Investigate Can a period-aware baseline outperform persistence on a fixed pseudo-periodic holdout?
Recorded license evidence is scoped to each source; it is not a blanket permission or a verification result.