learning-commons-evaluators 0.1.0

Python SDK for Learning Commons educational evaluators

learning-commons-evaluators (Python)

Python SDK for Learning Commons educational text evaluators. Evaluators call LLMs via LangChain, return structured Pydantic results, and share a common configuration and error-handling model.

Installation

pip install learning-commons-evaluators

Requires Python 3.10+. Provider API keys are passed in at runtime (not bundled with the package).

Quick start

import logging
from learning_commons_evaluators import (
    ConventionalityEvaluator,
    ConventionalityEvaluationInput,
    GoogleLLMProviderConfig,
    create_config_no_telemetry,
)

logging.basicConfig(level=logging.INFO)

config = create_config_no_telemetry(
    google_llm_provider_config=GoogleLLMProviderConfig(api_key="your-google-key"),
)

evaluator = ConventionalityEvaluator(config)
result = evaluator.evaluate_sync(
    ConventionalityEvaluationInput(text="The cat's out of the bag now.", grade=5)
)

print(result.answer.label)        # e.g. "Moderately complex"
print(result.explanation.summary) # Reasoning for the score

Docs

For full implementation details, check out the Python SDK docs.

More resources

• Local development – Local setup, testing, and development
• Evaluators — Shipped evaluators with their inputs, outputs, and evaluation settings
• Running evaluations — Sync / async usage and per-call settings overrides
• Results — EvaluationResult shape and metadata
• Configuration — Provider configs, EvaluatorConfig, evaluation settings, logging
• Error handling — Exception hierarchy, retries, and sanitization

License

MIT