Research, software, data and methods. Inspect the context, compare your options, and take a useful next step.
Executable Science local pilot drafts
Investigate four step sizes for y'=-10y using the known amplification factor 1-10h, including stable, boundary, and growing discrete trajectories.
Investigate When does a decaying continuous solution produce a non-decaying numerical trajectory?
SciPy documentation
A multiple-testing reference for adjusted p-values using Benjamini–Hochberg and Benjamini–Yekutieli procedures.
Investigate How do the two procedures differ on fixed p-value vectors and simulated dependent null tests?
Executable Science local pilot drafts
Compare unadjusted and Bonferroni thresholds on fixed-size families of synthetic uniform null p-values, with analytical familywise probabilities.
Investigate How does applying twenty tests change false-positive risk under a declared independent-null model?
Upstream project repository
Common filesystem interfaces and utilities for Python data-access libraries.
Investigate Do a selected local filesystem interface's seek and range reads match direct byte slicing?
Executable Science local pilot drafts
Sweep sixteen step sizes for forward and central differentiation of exp(x) at zero, exposing truncation and rounding behavior against the analytical derivative.
Investigate How does derivative error change across a fixed grid of shrinking step sizes?
Upstream project repository
Finite-volume partial differential equation solving for transport and related numerical models.
Investigate How do grid and time-step refinement affect error in a one-dimensional diffusion problem?
Upstream project repository
Attention-kernel implementations for comparing memory-efficient computation with a directly specified attention reference.
Investigate Does one supported attention kernel match dense scaled dot-product attention across mask and sequence-length edge cases?
Upstream project repository
Neural-network tooling built around JAX for investigating explicit state, transformations and model composition.
Investigate Does a tiny stateful model preserve its declared parameter/state boundary through a training update and checkpoint?
UCI Machine Learning Repository
Weather and fire observations for studying skewed environmental regression targets.
Investigate Does log-target modeling improve forest-fire prediction on a frozen spatial-temporal evaluation?
UCI Machine Learning Repository
Remote-sensing descriptors for examining forest-type identification across data groups.
Investigate How does forest-type accuracy change when location or acquisition groups are held out?
UCI Machine Learning Repository
Soybean cultivar observations across harvests for examining agricultural domain shift.
Investigate How does cultivar identification change when one harvest is withheld entirely?
UCI Machine Learning Repository
Gas-sensor responses over acquisition batches for investigating calibration drift.
Investigate How much does gas classification deteriorate when later acquisition batches are held out?
Recorded license evidence is scoped to each source; it is not a blanket permission or a verification result.