neurodesign-plus 2.0

Trial-aware fMRI experimental design optimisation for flat one-event and multi-event conceptual trials.

neurodesign-plus is the install distribution. neurodesign is the Python import package.

Version 2.0 keeps the public classes Experiment, Design, and Optimisation, but the active teaching surface is now trial-aware:

• conceptual trials can contain one or more modeled events

• flat order inputs are shorthand for one-event conceptual trials

• fixed workflows use complete trial templates

• probabilistic workflows sample complete templates for n_conceptual_trials

• timing is separated into event, within-trial, between-trial, and optional rest roles

• requested specifications, normalized rules, and realized timing arrays remain distinct

Version 2.0 is intentionally source-breaking. For legacy-to-current mapping, use the migration guide.

User Guide

• Installation
  • End-User Installation From PyPI
  • Local Source Installation
  • Editable Development Installation
  • Test And Notebook Dependencies
  • Documentation Dependencies
  • Version-2 Migration
• Getting Started
  • Public classes
  • Timing model
  • Minimal flat one-event example
  • Fixed multi-event example
  • Probabilistic complete-template sampling
  • Requested, normalized, and realized timing
  • Selector scope
  • Duration grammar
  • Counts and scoring
  • Convergence
  • Recommended workflow
  • Reproducibility
  • Next steps
• Migration Guide
• Migration To 2.0
  • Core Model
  • Current Public Construction Modes
  • Current Timing API
  • Legacy Name Mapping
  • Duration Grammar
  • Requested Versus Realized Timing
  • Optimization Workflow
  • Deterministic RNG
  • Convergence Semantics
  • Case 10
• Neurodesign-Plus 2.0 Metrics Guide
  • Scope
  • Conceptual Trials Versus Flattened Events
  • Design-Matrix Dependence
  • Event-Axis Metrics
  • Raw Versus Normalized Values
  • Weighted Objective
  • Current Flat Example
  • Current Optimisation Example
  • Convergence
• Technical Changes In 2.0
  • Authoritative Implementation
  • Trial-Aware Normalization
  • Timing-Rule Parsing
  • Requested, Normalized, And Realized State
  • Bounded Distribution Semantics
  • RNG Architecture
  • Schedule Representation
  • Trial-Boundary-Aware Optimization
  • Public Selection Workflow
  • Patience-Based Early Stopping
  • Reports And Export Payloads
  • Validation Runner

Reference

• API Reference
  • Core Classes
  • Utility Modules
• Tutorials
  • Maintained Notebook Inventory
  • Workflow Rules Taught Consistently
• Changelog
  • [2.0.0] - 2026-07-18
  • [1.0.2] - 2026-03-25
  • [1.0.1] - 2026-02-16
  • [1.0.0] - 2026-02-11
  • [0.2.02] - 2018-08-24
  • [0.2.01] - 2018-02-09
  • [0.2.00] - 2018-02-09
  • [0.1.13] - 2018-02-06
  • [0.1.12] - 2018-02-06
  • [0.1.08]-11 - 2018-02-06
  • [0.1.07] - 2018-02-05
  • [0.1.06] - 2018-11-22
  • [0.1.05] - 2017-11-22
  • [0.1.04] - 2017-08-30
  • [0.1.03] - 2017-08-09
  • [0.1.02] - 2017-07-31
  • [0.1.01] - 2017-07-03
  • [0.1.00] - 2017-04-26
  • [0.0.28] - 2017-02-20
  • [0.0.27] - 2017-02-20
  • [0.0.26] - 2017-02-14
  • [0.0.25] - 2017-02-06
  • [0.0.24] - 2017-02-02
  • [0.0.23] - 2017-02-01
  • [0.0.22] - 2017-02-01
  • [0.0.21] - 2017-02-01
  • [0.0.20] - 2017-02-01
  • [0.0.19] - 2017-02-01
  • [0.0.18] - 2017-02-01
  • [0.0.17] - 2017-02-01
  • [0.0.16] - 2017-02-01
  • [0.0.15] - 2017-01-31
  • [0.0.14] - 2017-01-19
  • [0.0.13] - 2017-01-18
  • [0.0.12] - 2017-01-18
  • [0.0.11] - 2017-01-17
  • [0.0.10] - 2017-01-10
  • [0.0.9] - 2016-12-19
  • [0.0.8] - 2016-12-19
  • [0.0.7] - 2016-12-09
  • [0.0.6] - 2016-11-18
  • [0.0.2] - 2016-11-18