On-Demand Feature View Examples

Using Request Data

from tecton import RequestSource, on_demand_feature_view
from tecton.types import Float64, Field, Bool

request_schema = [Field("amt", Float64)]
transaction_request = RequestSource(schema=request_schema)
output_schema = [Field("transaction_amount_is_high", Bool)]

# An example of an on-demand feature view that depends only on a request source.
@on_demand_feature_view(
    sources=[transaction_request],
    mode="python",
    schema=output_schema,
    description="The transaction amount is higher than $100.",
)
def transaction_amount_is_high(transaction_request):
    return {"transaction_amount_is_high": transaction_request["amt"] > 100}

Using Feature View Dependencies

from tecton import RequestSource, on_demand_feature_view
from tecton.types import String, Timestamp, Float64, Field, Bool
from fraud.features.stream_features.user_transaction_amount_metrics import (
    user_transaction_amount_metrics,
)

request_schema = [Field("amt", Float64)]
transaction_request = RequestSource(schema=request_schema)
output_schema = [Field("transaction_amount_is_higher_than_average", Bool)]


@on_demand_feature_view(
    sources=[transaction_request, user_transaction_amount_metrics],
    mode="python",
    schema=output_schema,
    description="The transaction amount is higher than the 1 day average.",
)
def transaction_amount_is_higher_than_average(transaction_request, user_transaction_amount_metrics):
    amount_mean = (
        0
        if user_transaction_amount_metrics["amt_mean_1d_10m"] is None
        else user_transaction_amount_metrics["amt_mean_1d_10m"]
    )
    return {"transaction_amount_is_higher_than_average": transaction_request["amt"] > amount_mean}

Embedding Cosine Similarity

Cosine similarity between embeddings is a common and powerful cross feature that can be implemented with On-Demand Feature Views. In this example, the feature view is computing the similarity between a pre-computed user embedding and a query embedding passed in the request to the Tecton Feature Service.

from tecton import on_demand_feature_view
from tecton.types import Field, Float64
from ads.features.feature_tables.user_embeddings import user_embeddings
from ads.features.feature_tables.ad_embeddings import ad_embeddings
import pandas

output_schema = [Field("cosine_similarity", Float64)]


@on_demand_feature_view(
    sources=[ad_embeddings, user_embeddings],
    mode="python",
    schema=output_schema,
    owner="jake@tecton.ai",
    tags={"release": "production"},
    description="Computes the cosine similarity between a precomputed ad embedding and a precomputed user embedding.",
)
def user_ad_embedding_similarity(ad_embedding, user_embedding):
    import numpy as np
    from numpy.linalg import norm

    def cosine_similarity(a: np.ndarray, b: np.ndarray):
        # Handle the case where one or both entities do not have a precomputed embedding.
        if a is None or b is None:
            return -1.0

        return np.dot(a, b) / (norm(a) * norm(b))

    result = {}
    result["cosine_similarity"] = cosine_similarity(
        user_embedding["user_embedding"], ad_embedding["ad_embedding"]
    ).astype("float64")
    return result