kaimei

Official Python client for the Kaimei explainability API.

Kaimei (解明) — to clarify, elucidate, reveal through investigation — turns trained black-box models into something you can inspect: feature-level attributions, global behaviour, learned rules, robustness, and drift.

from kaimei import Kaimei

client = Kaimei(api_key="km_live_...")

explanation = client.explanations.create(
    model="credit-risk-v3",
    prediction_id="pred_8f2a19",
    input={
        "income": 72000,
        "credit_score": 681,
        "utilization": 0.41,
    },
)

print(explanation.output)
print(explanation.top_features)
# 0.81 decline  |  credit_score -0.22

The same call as POST https://api.kaimei.ai/v1/explanations. See the product site for the language-agnostic REST examples.

Install

pip install kaimei

For a local API (the kaimei-api service in the KaimeiAI/kaimei repo):

client = Kaimei(api_key="km_test_...", base_url="http://localhost:8000")

KAIMEI_API_KEY and KAIMEI_BASE_URL are read from the environment when arguments are omitted.

What you can do

# Register a model (or point Kaimei at a remote inference URL / built-in demo)
client.models.create(
    name="Credit risk v3",
    slug="credit-risk-v3",
    task="binary_classification",
    features=[
        {"name": "income", "type": "numeric"},
        {"name": "credit_score", "type": "integer"},
        {"name": "utilization", "type": "numeric"},
    ],
    inference={"type": "builtin", "name": "credit-risk"},
)

# Log a prediction your own stack already made
client.predictions.create(model="credit-risk-v3", input={...}, output={"score": 0.81})

# Explain one input (Shapley by default; LIME or pre-computed attributions also accepted)
client.explanations.create(model="credit-risk-v3", input={...}, method="shapley")

# Study the model as a whole, then read the insights
analysis = client.analyses.create(model="credit-risk-v3", type="full")
client.analyses.wait(analysis.id)
insights = client.insights.list(model="credit-risk-v3")

Dashboard sign-in (session token) is also supported via client.auth.login(...) / client.auth.signup(...).

Resources

Errors raise kaimei.KaimeiError subclasses (AuthenticationError, NotFoundError, …) and expose .code, .type, and .status_code.

License

Apache 2.0. Copyright Kaimei AI.