gpcam

Plugin gpcam from marketplace gpcam in lbl-camera/gpCAM. Commands read from the repo's manifest — 9 days ago.

Skills (11)

• acquisition-functions

  Use when choosing or writing an acquisition function for gpCAM — selecting a built-in criterion for exploration, optimization, batch, or multi-task runs, or implementing a custom callable for UCB/LCB, minimization EI, constraints, radical gradient, or threshold finding.

• cost-functions

  Use when modeling the real expense of moving between gpCAM measurement points — motor travel time, settling, directional costs, sample damage, beam time, or zone-based penalties.

• experiment-designer

  Use for end-to-end autonomous experiment design with gpCAM. Translates a scientist's description of their measurement into a complete, runnable gpCAM script — useful for replacing raster scans with adaptive sampling, peak-finding, or parameter optimization.

• gp2scale-advanced

  Use for large-scale gpCAM experiments (>10k points up to millions) using sparse compactly-supported kernels and Dask distributed computing for exact GP computation at scale.

• kernel-designer

  Use when designing or composing custom kernel (covariance) functions for gpCAM — smoothness, periodicity, symmetry, anisotropy, non-stationary amplitude/length-scale kernels (`non_stat_kernel`, Gibbs), non-Euclidean inputs, or neural-network-warped deep kernels (`Network`).

• multi-task-advanced

  Use for multi-output, vector-valued, or function-valued gpCAM experiments using GPOptimizer with several output columns and multi_task_mode (or migrating legacy fvGPOptimizer code) — useful when a single measurement returns multiple correlated quantities (e.g., spectra, multi-channel detectors).

• noise-functions

  Use when modeling position-dependent, heteroscedastic, or otherwise structured noise in gpCAM — e.g., detector characteristics, count-rate-dependent variance, or non-uniform measurement uncertainty.

• prior-mean-functions

  Use when encoding known physics, theoretical models, or expected trends as prior mean functions for gpCAM — useful when there's a baseline expectation or physical model the GP should regress against rather than the default constant data-mean prior (mean(y_data)).

• transformed-optimizers-advanced

  Use when observations are constrained — strictly positive (intensities, rates, concentrations) or bounded in [a, b] (fractions, probabilities, yields). LogGPOptimizer and LogitGPOptimizer fit a GP on transformed observations and push the posterior back through the inverse link, so predictions and cr

• troubleshooting

  Use when a gpCAM script crashes or misbehaves — Cholesky/non-positive-definite errors, singular covariance, hyperparameter bounds mismatches, NaNs, kernel signature errors, Dask hangs, pickling failures, or an ask/tell loop that keeps picking the same point. Maps the actual error text to the cause a

• uncertainty-calibration

  Use to validate that a gpCAM posterior's error bars are trustworthy before publishing them or letting an autonomous loop act on them — scoring rules (NLPD, CRPS, MSLL), calibration/coverage curves, PICP and interval width, and what to do when the model is over- or under-confident.

Similar plugins