LookoutEquipmentDataset

class src.lookoutequipment.dataset.LookoutEquipmentDataset(dataset_name, access_role_arn, component_fields_map=None, component_root_dir=None)

A class to manage Lookout for Equipment datasets

Attributes

components_list

list of components part of the schema of this dataset

dataset_name

string with the name given to the dataset

dataset_schema

string with a JSON-formatted string describing the data schema the dataset must conform to

ingestion_job_id

string the ID of the data ingestion job

ingestion_job_response

string with a JSON-formatted string describing the response details of a data ingestion job.

role_arn

string containing the role ARN necessary to access the S3 location where the datasets are stored

schema

dict dictionnary containing the schema of this dataset if it was already created in Lookout for Equipment

Methods

__init__(dataset_name, access_role_arn[, …])

Create a new instance to configure all the attributes necessary to manage a Lookout for Equipment dataset.

create()

Creates a Lookout for Equipment dataset

delete([force_delete])

Deletes the dataset

get_component_field_map(component)

ingest_data(bucket, prefix[, wait, sleep_time])

Ingest data from an S3 location into the dataset

list_models()

List all the models trained with this dataset

poll_data_ingestion([sleep_time])

This function polls the data ingestion describe API and prints a status until the ingestion is done.
__init__(dataset_name, access_role_arn, component_fields_map=None, component_root_dir=None)

Create a new instance to configure all the attributes necessary to manage a Lookout for Equipment dataset.

Parameters

  • dataset_name (string) – the name of the dataset to manage

  • component_fields_map (string) – the mapping of the different fields associated to this dataset. Either component_root_dir or component_fields_map must be provided. Defaults to None.

  • component_root_dir (string) – the root location where the sensor data are stored. Either component_root_dir or component_fields_map must be provided. Defaults to None. Can be a local folder or an S3 location.

  • access_role_arn (string) – the ARN of a role that will allow Lookout for Equipment to read data from the data source bucket on S3

create()

Creates a Lookout for Equipment dataset

Returns

Response of the create dataset API

Return type

string

delete(force_delete=True)

Deletes the dataset

Parameters

force_delete (boolean) – if set to True, also delete all the models that are using this dataset before deleting it. Otherwise, this method will list the attached models (Default: True)

get_component_field_map(component)
ingest_data(bucket, prefix, wait=False, sleep_time=60)

Ingest data from an S3 location into the dataset

Parameters

  • bucket (string) – Bucket name where the data to ingest are located

  • prefix (string) – Actual location inside the aforementioned bucket

  • wait (Boolean) – If True, this function will wait for the ingestion to finish (default to False)

  • sleep_time (integer) – how many seconds should we wait before polling again when the wait parameter is True (default: 60)

Returns

Response of the start ingestion job API call (if wait is False) or of the actual finished ingestion job (if wait is True)

Return type

string

list_models()

List all the models trained with this dataset

Returns

A list with the names of every models trained with this dataset

Return type

list of strings

poll_data_ingestion(sleep_time=60)

This function polls the data ingestion describe API and prints a status until the ingestion is done.

Parameters

sleep_time (integer) – How many seconds should we wait before polling again (default: 60)