Service

Flask application to serve Machine Learning models

service.features()

Model features

Get the model accepted features. This includes feature inportance if the model allows it.

service.health_check()

service.info()

Model information

Get the model information: metadata, type, classifier, etc.

service.predict()

Make preditcions and explain them

Model inference using input data. This is the main function.

URL Params:

proba (int):

1 in order to compute probabilities for classification models or 0 to return predicted class (classification) or value (regression). Default 0.

explain (int):

1 in order to compute moeldel explanations for the predicted value. This will return a status 500 when the model does not support explanations. Default 0.

Payload:

JSON string that can take two forms:

The first, the payload is a record or a list of records with one value per feature. This will be directly interpreted as the input for the model.

The second, the payload is a dictionary with 1 or 2 elements. The key “_data” is mandatory because this will be the input for the model and its format is expected to be a record or a list of records. On the other hand the key “_samples” (optional) will be used to obtain different explanations (see explain())

service.preprocess()

Preporcess input data

Get the preprocessed version of the input data. If the model does not include preprocessing steps, this method will return the same data as the input.

service.readiness_check()

service.service_info()

Service information

Get information about the service: up-time, varsion of the template, name of the served model, etc.