Inference logger#
Creation#
InferenceLogger#
hsml.inference_logger.InferenceLogger(kafka_topic={}, mode="ALL", **kwargs)
Configuration of an inference logger for a predictor.
Arguments
- kafka_topic
Optional[Union[hsml.kafka_topic.KafkaTopic, dict]]
: Kafka topic to send the inference logs to. By default, a new Kafka topic is configured. - mode
Optional[str]
: Inference logging mode. (e.g.,NONE
,ALL
,PREDICTIONS
, orMODEL_INPUTS
). By default,ALL
inference logs are sent.
Returns
InferenceLogger
. Configuration of an inference logger.
Retrieval#
predictor.inference_logger#
Inference loggers can be accessed from the predictor metadata objects.
predictor.inference_logger
Predictors can be found in the deployment metadata objects (see Predictor Reference). To retrieve a deployment, see the Deployment Reference.
Properties#
kafka_topic#
Kafka topic to send the inference logs to.
mode#
Inference logging mode ("NONE", "ALL", "PREDICTIONS", or "MODEL_INPUTS").
Methods#
describe#
InferenceLogger.describe()
Print a description of the inference logger
to_dict#
InferenceLogger.to_dict()