ke

A framework for building Knowledge Evaluation systems — evaluating the outputs of information-extraction systems. OCR is treated as the noisiest special case of a general problem, so the core is source-agnostic and the OCR pieces are optional.

import ke

ke.score("hello wrld", "hello world")  # -> Score(value=0.0909..., metric='cer')
ke.score("hello wrld", "hello world", metric="wer").value  # 0.5
ke.evaluate(
    [("ct", "cat"), ("dg", "dog")], metric="cer"
).aggregate  # 0.333... (global CER)

What it does

Evaluating an extraction splits along two axes — is there a gold answer (reference-based) or not (reference-free), and are we scoring one item or a whole corpus. ke gives you both halves through two facades over one shared typed schema:

• score() / evaluate() — reference-based: compare against gold, one item or a corpus, the metric chosen by output type (string → CER/WER, record → field-F1), aggregated correctly (global error-rate accumulation, micro-F1; never a naive mean) with optional per-slice cuts.
• estimate_quality() — reference-free: gather signals → calibrate → validate → decide accept/flag/block, with no gold answer.

Everything swappable is a strategy injected with a smart default, so the simple call works out of the box and every layer stays replaceable.

Evaluate an OCR engine

The first concrete instance: measure OCR accuracy over a gold corpus. ke consumes ocracy's normalized OcrResult, so it can benchmark any of its ~16 engines — or any image -> OcrResult callable of your own.

import ke.ocr

gold = {
    "inv-1": {
        "image": "scan.png",
        "reference_text": "INVOICE 2024",
        "slice": "invoices",
    }
}
report = ke.ocr.evaluate_ocr(
    "ocrmac",
    gold,
    metric="cer",
    normalize=["lower", "collapse_whitespace"],
    persist=True,
)
report.aggregate  # corpus CER
report.per_slice  # CER per document slice
report.detail["per_item"]  # prediction, reference, score, confidence per document

Gold corpora, results, and runs persist to local dol stores under ~/.local/share/ke/.

Install

pip install ke            # lean, permissive core (dol, config2py, jiwer, rapidfuzz)
pip install "ke[ocr]"     # + the ocracy OCR fleet (install engines via ocracy extras)
pip install "ke[all]"     # + the permissive capability tiers (metrics, calibration, ...)

Heavier or copyleft/non-commercial libraries are never installed by default; see the extras in pyproject.toml. Some capabilities (e.g. the cost-weighted typed-graph metric and the ROVER consensus engine) are on the roadmap — see the tracking issue.

CLI

ke cer "hello wrld" "hello world"     # character error rate
ke wer "hello wrld" "hello world"     # word error rate
ke where                              # the local data folder
ke check tesseract                    # what an OCR engine needs to run

For contributors

The architecture, conventions, and the research behind the design are documented for agents and humans in AGENTS.md, the dev skills under skills/, and the research reports under misc/docs/.