Regression

How to log your model schema for regression models

Regression Model Overview

Regression models have a continuous, numeric output. (Examples: click-through rates, sales forecasting, customer lifetime value, ETA models, etc.)

Performance Metrics

MAPE, MAE, RMSE, MSE, R-Squared, Mean Error

Allowed Metric Families: Regression

Click here for all valid model types and metric combinations.

Regression Code Example

Example Row

price (float)

pos_approved (bool)

zip_code

age

prediction_score

actual_score

prediction_ts

88.5

False

12345

100

90

1671572541

# feature & tag columns can be optionally defined with typing:
tag_columns = TypedColumns(
    inferred=["name"],
    to_int=["zip_code", "age"]
)

# Declare the schema of the dataframe you're sending (feature columns, predictions, timestamp, actuals) 
schema = Schema(
    prediction_id_column_name="prediction_id",
    timestamp_column_name="prediction_ts",
    prediction_score_column_name="prediction_score",
    actual_score_column_name="actual_score",
    feature_column_names=["price", "pos_approved"],
    tag_column_names=tag_columns,
)
# Log the dataframe with the schema mapping
response = client.log(
    model_id='sample-model-1', 
    model_version='v1', 
    model_type=ModelTypes.REGRESSION,
    metrics_validation=[Metrics.REGRESSION],
    environment=Environments.PRODUCTION,
    dataframe=test_dataframe,
    schema=schema
)

# features & tags can be optionally defined with typing
features = {
    'score': '88.5',
    'pos_approved': TypedValue(value=False, type=ArizeTypes.INT)
}  
    
tags = {
    'zip_code': '12345'
    'age': '25'
}

response = client.log(
    model_id='sample-model-1', 
    model_version='v1', 
    model_type=ModelTypes.REGRESSION,
    environment=Environments.PRODUCTION,
    features = features, 
    tags=tags
    prediction_label=100,
    actual_label=90
)

Quick Definitions

Prediction Label: The numeric value of the prediction (float | int)

Actual Label: The numeric value of the actual (float | int)

PreviousMulti-Class Classification NextLending

Last updated 1 month ago

# feature & tag columns can be optionally defined with typing: tag_columns = TypedColumns( inferred=["name"], to_int=["zip_code", "age"] ) # Declare the schema of the dataframe you're sending (feature columns, predictions, timestamp, actuals) schema = Schema( prediction_id_column_name="prediction_id", timestamp_column_name="prediction_ts", prediction_score_column_name="prediction_score", actual_score_column_name="actual_score", feature_column_names=["price", "pos_approved"], tag_column_names=tag_columns, ) # Log the dataframe with the schema mapping response = client.log( model_id='sample-model-1', model_version='v1', model_type=ModelTypes.REGRESSION, metrics_validation=[Metrics.REGRESSION], environment=Environments.PRODUCTION, dataframe=test_dataframe, schema=schema )

# features & tags can be optionally defined with typing features = { 'score': '88.5', 'pos_approved': TypedValue(value=False, type=ArizeTypes.INT) } tags = { 'zip_code': '12345' 'age': '25' } response = client.log( model_id='sample-model-1', model_version='v1', model_type=ModelTypes.REGRESSION, environment=Environments.PRODUCTION, features = features, tags=tags prediction_label=100, actual_label=90 )