About Embedded Python Execution

Embedded Python execution is a feature of Oracle Machine Learning for Python (OML4Py) that allows users to invoke user-defined Python functions directly in a database instance.

See Embedded Python Execution in Oracle Machine Learning for Python User’s Guide for more information.

To demonstrate how to use embedded execution, the following example prepares a spatial regression model. This model is used in the subsequent topics in this chapter which describe how to invoke a Python function using embedded execution, including creating a function that uses that model to make predictions, and then using that function for embedded execution from Python, SQL and REST.

The example steps are as follows:

  1. Defines the regressor model using the block_groups SpatialDataFrame and SpatialErrorRegressor.
  2. Creates a Spatial Pipeline with a pre-processing step to standardize the data and the regressor as the last step.
  3. Trains the model using MEDIAN_INCOME as the target variable. 
from oraclesai.preprocessing import spatial_train_test_split
from oraclesai.weights import KNNWeightsDefinition
from oraclesai.regression import SpatialErrorRegressor
from oraclesai.pipeline import SpatialPipeline
from sklearn.preprocessing import StandardScaler
 
# Define variables
X = block_groups[["MEDIAN_INCOME", "MEAN_AGE", "HOUSE_VALUE", "geometry"]]
 
# Define training and test sets
X_train, X_test, _, _, _, _ = spatial_train_test_split(X, y="MEDIAN_INCOME", test_size=0.2, random_state=32)
 
# Create instance of SpatialErrorRegressor
spatial_error_model = SpatialErrorRegressor(spatial_weights_definition=KNNWeightsDefinition(k=5))
 
# Add the model into a Spatial Pipeline along with a pre-processing step
spatial_error_pipeline = SpatialPipeline([("scaler", StandardScaler()), ("spatial_error", spatial_error_model)])
 
# Train the model with MEDIAN_INCOME as the target variable
spatial_error_pipeline.fit(X_train, "MEDIAN_INCOME")

Once the model is trained, save the model into an OML4Py datastore.

oml.ds.save({'spatial_error': spatial_error_pipeline}, 
    'spatial_error_ds', description='some description', 
    overwrite=True)