Learning objectives
- Package an analysis so another researcher can rerun and test it.
- Explain pure functions, pinned environments, tests, notebooks, and manifests with the formal object.
- Separate evidence, transformation, inference, and limitation.
Terms Developed in Context
These terms form the lecture's reasoning chain from observable through model to a bounded physical conclusion.
- Pure function
A computation whose output depends only on explicit inputs and that produces no hidden side effects. In this lecture it supports the evidence-to-model chain for pure functions, pinned environments, tests, notebooks, and manifests.
Use and relationship: Read it together with Notebook execution order and Pinned environment. The governing relation \(f(data;code,dependencies,seed)->result+metadata\) shows where the concept enters quantitatively; the student should state whether its value is measured, calibrated, adopted, or inferred.
Boundary: Reproducibility is testable when expected inputs and outputs are explicit. This caveat prevents the term from being treated as a unique explanation or a context-free number.
- Pinned environment
A dependency specification fixing versions sufficiently to recreate the software context. In this lecture it supports the evidence-to-model chain for pure functions, pinned environments, tests, notebooks, and manifests.
Use and relationship: Read it together with Pure function and Invariant test. The governing relation \(f(data;code,dependencies,seed)->result+metadata\) shows where the concept enters quantitatively; the student should state whether its value is measured, calibrated, adopted, or inferred.
Boundary: Reproducibility is testable when expected inputs and outputs are explicit. This caveat prevents the term from being treated as a unique explanation or a context-free number.
- Invariant test
A check of a property that should remain true across valid inputs or refactoring. In this lecture it supports the evidence-to-model chain for pure functions, pinned environments, tests, notebooks, and manifests.
Use and relationship: Read it together with Pinned environment and Random seed. The governing relation \(f(data;code,dependencies,seed)->result+metadata\) shows where the concept enters quantitatively; the student should state whether its value is measured, calibrated, adopted, or inferred.
Boundary: Reproducibility is testable when expected inputs and outputs are explicit. This caveat prevents the term from being treated as a unique explanation or a context-free number.
- Random seed
A recorded initializer that makes a pseudorandom computational path repeatable. In this lecture it supports the evidence-to-model chain for pure functions, pinned environments, tests, notebooks, and manifests.
Use and relationship: Read it together with Invariant test and Notebook execution order. The governing relation \(f(data;code,dependencies,seed)->result+metadata\) shows where the concept enters quantitatively; the student should state whether its value is measured, calibrated, adopted, or inferred.
Boundary: Reproducibility is testable when expected inputs and outputs are explicit. This caveat prevents the term from being treated as a unique explanation or a context-free number.
- Notebook execution order
The dependency sequence among interactive cells, which must be reproducible from a clean state. In this lecture it supports the evidence-to-model chain for pure functions, pinned environments, tests, notebooks, and manifests.
Use and relationship: Read it together with Random seed and Pure function. The governing relation \(f(data;code,dependencies,seed)->result+metadata\) shows where the concept enters quantitatively; the student should state whether its value is measured, calibrated, adopted, or inferred.
Boundary: Reproducibility is testable when expected inputs and outputs are explicit. This caveat prevents the term from being treated as a unique explanation or a context-free number.
Method in context
Reproducibility is engineered through explicit inputs, deterministic transformations, tests, and a discoverable environment. Ask what a second researcher needs to reproduce the claim: raw input, processing state, software or database version, and inferential step.
Derivation and assumptions
Define every symbol and explain the failure mode. A method becomes scientific when assumptions are visible enough to be tested.
Worked example
A deterministic function with a fixed seed and invariant test exposes hidden state.
Reproducibility is testable when expected inputs and outputs are explicit. Require an independent recomputation and one perturbation.
Misconceptions and study guidance
Do not confuse a data service with a measurement, a catalog identifier with physical identity, a formal error with total uncertainty, or a polished figure with a validated inference. Reconstruct the evidence chain before interpreting.
Verified reading
Astropy citation guidance; Python Packaging User Guide, reproducible environments.
Use the exact documentation anchor in the reference log and record any access mismatch.