Skip to content

tutorials.md

Latest commit

 

History

History
35 lines (28 loc) · 2.76 KB

tutorials.md

File metadata and controls

35 lines (28 loc) · 2.76 KB
Raw

Tutorials

The maintained tutorial surface contains 11 authored notebooks. They are tracked directly under tutorials/ and executed from clean kernels during validation.

Maintained Notebook Inventory

Notebook Focus
tutorials/tutorial_1-neurodesign_base_overview.ipynb End-to-end overview: manual designs, crossover/mutation, and GA vs. simulation vs. random-baseline optimization
tutorials/tutorial_2-comparing_designs_across_experiments.ipynb When raw Fe/Fd are comparable across different Experiment setups (same whitening matrix), and when they aren't
tutorials/tutorial_3-progressive_experiment_building.ipynb Progressive Case 1-10 walkthrough, building from a minimal event-related design up to full branching-trial complexity
tutorials/base_functions/tutorial_base-comparing_designs.ipynb Comparing five hand-built designs (cyclic, blocked, crossover, mutation) on Fe/Fd/Fc/Ff
tutorials/base_functions/tutorial_base-designing_scoring_and_optimizing.ipynb Comparing GA, simulation, and random-baseline optimization via efficiency scores and simulated statistical power
tutorials/base_functions/tutorial_base-discovering_best_design.ipynb Genetic-algorithm design discovery: comparing hand-built designs, then searching for a better one automatically
tutorials/base_functions/tutorial_base-optimizing_and_reporting.ipynb Automatic GA optimization plus PDF report generation
tutorials/new_functions/tutorial_new-event_and_trial_intervals.ipynb Transition-specific ITIs via event_transition_interval
tutorials/new_functions/tutorial_new-fixed_ordering.ipynb Fixing the stimulus order with order= so optimization only searches over timing
tutorials/new_functions/tutorial_new-probabilistic_ordering.ipynb Template-based probabilistic ordering with order_keys / order_probabilities
tutorials/new_functions/tutorial_new-variable_event_durations.ipynb Per-stimulus variable event durations via event_durations

Workflow Rules Taught Consistently

The maintained notebooks teach these public version-2 rules consistently:

  • use Experiment to declare the requested specification
  • use create_design(...) or create_manual_design(...) for no-search workflows
  • use Optimisation only when search is needed
  • retrieve the authoritative design with selected_design(0) after optimise()
  • generate reports and exports from that selected design workflow
  • distinguish n_conceptual_trials from flattened n_events
  • treat Fe and Fd as realized design-matrix metrics
  • treat Ff and Fc as flattened event-axis metrics

Notebook execution is automated through validation.execute_notebooks and covered by the aggregate validation runner in validation.run_all_validation_workflows.