Dataset
Summary​
Persisted data consisting of entity & request keys, timestamps, and calculated features. Datasets are associated with either a FeatureService or FeatureView.Â
There are 2 types of Datasets: Saved and Logged.
Â
Saved Datasets are generated manually when calling
.start_dataset_job()
on Tecton DataFrame, ie:Â Logged Datasets are generated automatically when declaring a FeatureService with LoggingConfig, and the data is continuously added to it when requesting online data from the FeatureService.
Â
To get an existing Dataset, call
workspace.get_dataset()
.Example
data_frame = get_features_for_events(my_spine)
data_frame.start_dataset_job(dataset_name='my_training_dataset:V1')
Attributes​
Name | Data Type | Description |
---|---|---|
is_archived | bool | Boolean indicating if the dataset is archived. |
name | str | Name of the dataset. |
Methods​
Name | Description |
---|---|
__init__(...) | Initialize self. See help(type(self)) for accurate signature. |
to_dataframe(...) | Loads the data and returns it as TectonDataFrame |
to_dataframe(...)​
Loads the data and returns it as TectonDataFrameParameters
start_time
(Optional
[datetime.datetime
]) - The interval start time from when we want to retrieve the data. Default:None
end_time
(Optional
[datetime.datetime
]) - The interval end time until when we want to retrieve the data. Default:None
Returns
TectonDataFrame