Performance and comparisons
These treatments share a subject; their labels distinguish the questions they answer.
- Inspection and explicit computation — Library guide. What is already retained, what does a query compute, and what can I reuse across repeated queries or a batch?
- Keeping a selected mathematical view — Library guide. How do I create, query and close a linked session without losing the meaning of a selection?
- Concurrent polyhedral computations — Task recipe. How can I safely run polyhedral computations concurrently?
- Reproduce a benchmark report — Task recipe. How do I verify the published evidence, rerun a comparison and redraw its figures?
- Linked inspection sessions and browser execution — Implementation account. Which representations and algorithms underlie linked inspection sessions and browser execution, and why?
- C++ Ripser: ordinary Rips barcodes — Benchmark report. How does TamerOp compare with original C++ Ripser on larger complete-barcode requests?
- Continuous inspection: time from input to picture — Benchmark report. How long does a continuous control take to commit and draw its mathematical result on the declared browser fixtures?
- CubicalRipser: cubical barcodes — Benchmark report. How do TamerOp and CubicalRipser compare on large image and volume barcodes?
- PHAT: ordinary persistence — Benchmark report. What does the scoped comparison with PHAT establish?
- QPA: finite module algebra — Benchmark report. What does the scoped comparison with QPA establish?
- Ripser.py: Vietoris–Rips persistence — Benchmark report. What does the fixed comparison with Ripser.py establish for ordinary Rips barcodes, cocycles and landmark workflows?
- How the benchmarks were measured — Contributor guide. How do I measure a meaningful computation and report what the evidence supports?