quantum-corpus
Version 0.3.4 — RAG pipeline and corpus tools for quantum research copilot.
A local-only corpus pipeline that ingests quantum research material from multiple sources, applies PII redaction, splits records into train/val/test sets, builds a versioned SQLite corpus, and constructs a BM25 + optional semantic hybrid RAG index.
Installation
pip install quantum-corpus # core only (BM25 RAG, no ML deps)
pip install "quantum-corpus[advanced]" # + sentence-transformers + PyMuPDF
pip install "quantum-corpus[dev]" # + pytest
Core Modules
| Module | Description | Dependencies |
|---|---|---|
quantum_corpus.schema |
SQLite schema + source_identity | sqlite3 (stdlib) |
quantum_corpus.redact |
PII redaction (IBMid, keys, emails) | re (stdlib) |
quantum_corpus.split |
Train/val/test split by project | hashlib (stdlib) |
quantum_corpus.tokenize_count |
Token counting with HuggingFace tokenizers | tokenizers |
quantum_corpus.rag |
BM25 RAG index + schema-aware query expansion | — |
quantum_corpus.fusion |
BM25 + semantic hybrid fusion | quantum_corpus.rag, quantum_corpus.semantic |
quantum_corpus.structured |
Structured SQL query layer | sqlite3 (stdlib) |
quantum_corpus.answer |
Evidence-gated answer synthesis | quantum_corpus.rag, quantum_corpus.structured |
quantum_corpus.extract |
Ingest from repo dirs, IBM job zips, PDFs | fitz (optional) |
quantum_corpus.build |
End-to-end build orchestrator | all above |
Quick Start
from quantum_corpus import schema, rag
# Use existing corpus
db_path = schema.default_db_path() # or set TMT_QUANTUM_CORPUS_DB
idx = rag.RAGIndex.load(db_path)
# Query
hits = idx.query("ibm_fez backend job OTOC", k=5)
for hit in hits:
print(hit["snippet"][:120])
Environment Variables
| Variable | Default | Description |
|---|---|---|
TMT_QUANTUM_CORPUS_DB |
quantum_corpus.db |
Path to corpus SQLite DB |
TMT_QUANTUM_JOBS_DB |
— | Path to IBM job structured DB |
TMT_DEPLOY_MODE |
private-training |
private-training or public-demo |
Corpus Schema
Each record has:
id,source_type,project,subdomain,doc_id,textsplit(train/val/test),token_countsensitivity(public/internal/restricted/sensitive)risk_tier(public/standard/elevated/critical)source_identity(stable SHA-256 content hash)provenance_url,source_license
License
MIT. See LICENSE file.
1 maintainer
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distributions
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file quantum_corpus-0.3.4-py3-none-any.whl.
File metadata
- Download URL: quantum_corpus-0.3.4-py3-none-any.whl
- Upload date:
- Size: 89.3 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? Yes
- Uploaded via:
twine/7.0.0 CPython/3.13.14
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
5119e270263cae9931ff16f6da6c14acbeeb701ae4b0d27bbc2cbb6a733ecb89
|
|
| MD5 |
830fa27042e58239d8ea80cf00a6d385
|
|
| BLAKE2b-256 |
a782e2a6d34530d21572d8bf1ee278f6b6db5a816e686dfdf89046dc89c621bb
|
Provenance
The following attestation bundles were made for quantum_corpus-0.3.4-py3-none-any.whl:
Publisher:
pypi-publish.yml on quantumdynamics927-dotcom/tinymetatron
-
Statement:
-
Statement type:
https://in-toto.io/Statement/v1 -
Predicate type:
https://docs.pypi.org/attestations/publish/v1 -
Subject name:
quantum_corpus-0.3.4-py3-none-any.whl -
Subject digest:
5119e270263cae9931ff16f6da6c14acbeeb701ae4b0d27bbc2cbb6a733ecb89 - Sigstore transparency entry: 2315613114
- Sigstore integration time:
-
Permalink:
quantumdynamics927-dotcom/tinymetatron@250e3d9970a36734a430adaffc517e2b55954ebe -
Branch / Tag:
refs/tags/v0.3.4 - Owner: https://github.com/quantumdynamics927-dotcom
-
Access:
private
-
Token Issuer:
https://token.actions.githubusercontent.com -
Runner Environment:
github-hosted -
Publication workflow:
pypi-publish.yml@250e3d9970a36734a430adaffc517e2b55954ebe -
Trigger Event:
push
-
Statement type:
