Skip to main content

Betti — Topological-Shape Oracle 🕳️

Betti sells the shape of data. Given a point cloud it builds a Vietoris–Rips filtration and runs the standard GF(2) persistence reduction to read off the Betti numbers as a function of scale:

  • b₀ — connected components (how many clusters),
  • b₁ — loops / holes (1-cycles that don't fill in),
  • b₂ — voids / cavities (hollow 3-D regions),

plus the full persistence barcode/diagram and a bottleneck distance between two diagrams — a basis-free, noise-stable drift detector: ~0 when two clouds share a topology, clearly positive when the shape changed (a loop appeared, components merged, a cavity opened).

Built on oracle-core (AIMarket Protocol v2). Where Lumen scores who to trust and Percola finds when a graph collapses, Betti answers what shape the data has — structure no scalar summary statistic can express.

# Homology: 24 points on a ring → one loop (b1 = 1), one component (b0 = 1)
curl -s -X POST http://localhost:9313/ai-market/v2/invoke -H "Content-Type: application/json" \
  -d '{"capability_id":"betti.homology@v1","input":{"points":[[1,0],[0.97,0.26],[0.87,0.5],[0.71,0.71],[0.5,0.87],[0.26,0.97],[0,1],[-0.26,0.97],[-0.5,0.87],[-0.71,0.71],[-0.87,0.5],[-0.97,0.26],[-1,0],[-0.97,-0.26],[-0.87,-0.5],[-0.71,-0.71],[-0.5,-0.87],[-0.26,-0.97],[0,-1],[0.26,-0.97],[0.5,-0.87],[0.71,-0.71],[0.87,-0.5],[0.97,-0.26]],"max_scale":0.5}}'
Capability What Price
betti.homology@v1 b₀/b₁/b₂ at scale, Betti curve, persistence diagram $0.008
betti.distance@v1 bottleneck distance between two diagrams (topology drift) $0.004

The mathematics

  • Vietoris–Rips: vertices at scale 0; an edge when a pair distance ≤ ε; a triangle when all three edges exist; a tetrahedron when all four faces exist (needed for a true b₂). Each simplex's filtration value (birth) is its longest edge.
  • Reduction: sort simplices by (filtration, dimension), build the boundary matrix as GF(2) columns of face-index sets, reduce left-to-right by the low-row pivot. Each pivot pairs a birth simplex with a death simplex → a persistence interval [birth, death]; unpaired columns are essential (infinite) bars. b₀ is cross-checked with union–find.
  • Betti curve: b_k(ε) = number of dim-k bars alive at ε.
  • Bottleneck: exact matching via binary search on ε + Hopcroft–Karp/Kuhn perfect matching on the threshold bipartite graph, with the diagonal allowed. Complexity per feasibility check O(E·√V); V ≤ 2(|A|+|B|) after the diagram cap.

Hard caps (no silent truncation)

The protocol layer does not validate input, so the handlers do. A Rips complex is combinatorially explosive, so Betti caps points at 300 and total simplices at 150 000; if it clips, it says so in the response notes/capped fields rather than quietly dropping data. Missing/malformed points raise ValueError → {ok:false}.

Part of the AICOM oracle family. MIT.

Metadata

Release files for aimarket-oracle-betti 0.1.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for aimarket-oracle-betti 0.1.0
File Size Uploaded Provenance
aimarket_oracle_betti-0.1.0.tar.gz 15.7 kB Without provenance Details

Built distribution (wheel)

Table of built distributions (wheels) for aimarket-oracle-betti 0.1.0
File Interpreter ABI Platform
aimarket_oracle_betti-0.1.0-py3-none-any.whl Python 3 none any Details

Total release size: 29.2 kB

Release files / aimarket_oracle_betti-0.1.0.tar.gz

Download URL aimarket_oracle_betti-0.1.0.tar.gz
Size 15.7 kB
Tags Source
SHA-256 checksum
How to use checksums
6f4e09bd9540a97f4a8a2068f6b754b17df715c32d33337adf93c7eb7080d9ed
BLAKE2b-256 checksum
How to use checksums
52bcff69661778e9f8045d87f61fe6f6f4c0e4361742cc5a732136b1d4fae1c4
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.9.6

Release files / aimarket_oracle_betti-0.1.0-py3-none-any.whl

Download URL aimarket_oracle_betti-0.1.0-py3-none-any.whl
Size 13.5 kB
Tags Python 3
SHA-256 checksum
How to use checksums
787e779da956c603fe7d8e3d7cecd3fb2c9e259c69e3847bfae48036184761b9
BLAKE2b-256 checksum
How to use checksums
9863c677540f728eee59f2f7e6b1082b39fd6143131cc9b81946476979f06ec2
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.9.6

Security for aimarket-oracle-betti 0.1.0

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Release history Release notifications | RSS feed

This release

0.1.0 This release

2 release files No provenance

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page