rockfish.metrics

`metrics`

Classes

Functions

`session_length(dataset: LocalDataset, session_length_field: str = 'session_length') -> LocalDataset`

Returns a new dataset with the metadata of table and a field containing the session length.

Parameters:

Name	Type	Description	Default
`dataset`	`LocalDataset`	Input dataset.	required
`session_length_field`	`str`	New field name to hold session length.	`'session_length'`

Raises:

Type	Description
`rf.errors.TableMetadataMissingError`	When the dataset does not contain dataset metadata.

`count_all(dataset: LocalDataset, field: str, *, nlargest: Optional[int] = None) -> LocalDataset`

Returns a new table containing the distinct values for the specified field and a new field with the number times the value occurred. The new field is named field_count.

Parameters:

Name	Type	Description	Default
`dataset`	`LocalDataset`	Input dataset.	required
`field`	`str`	Field name to count.	required
`nlargest`	`Optional[int]`	Limit results to nlargest values.	`None`

`interarrivals(dataset: LocalDataset, time_field: str, interarrival_field: str = 'interarrival', unit: Precision = 's') -> LocalDataset`

Create a table containing the interarrival times.

Metadata fields are determined using the schema metadata.

The table will contain the metadata and the interarrival times. Each session has n-1 rows due to the first point not having a interarrival delta.

Parameters:

Name	Type	Description	Default
`dataset`	`LocalDataset`	Input dataset.	required
`time_field`	`str`	Field in dataset containing times.	required
`interarrival_field`	`str`	Field to be added with interarrival times.	`'interarrival'`

Raises:

Type	Description
`arrow.MetadataMissingError`	When table does not contain schema metadata.

`aggregate(dataset: LocalDataset, field: str, agg_func_name: AggregateMethod, group_fields: Optional[list[str]] = None, output_field: Optional[str] = None) -> LocalDataset`

Returns a new table containing the aggregate values in the specified field for each group. If group_fields == [], the aggregation will be applied to the entire dataset.

Parameters:

Name	Type	Description	Default
`dataset`	`LocalDataset`	Input dataset.	required
`field`	`str`	Field name to compute aggregation.	required
`agg_func_name`	`AggregateMethod`	Aggregation function to use. For numerical field: "sum", "mean", "min", "max", "variance", "stddev", "count", "count_distinct". For categorical field: "count", "count_distinct".	required
`group_fields`	`Optional[list[str]]`	Optional. The fields to group by. The precedence order for group fields is this group_fields > table_metadata.group_fields > table_metadata.metadata. Default is None.	`None`
`output_field`	`Optional[str]`	Optional. Name of the output field. Default is None. The output will be named field_agg_func_name.	`None`

`transitions_within_sessions(dataset: LocalDataset, field: str, metadata_fields: Optional[list[str]] = None, k_gram: Optional[int] = 2, collapse: bool = False) -> LocalDataset`

Creates a new dataset that lists session keys to define sessions and the state transitions for a given stateful field within each session.

Parameters:

Name	Type	Description	Default
`dataset`	`LocalDataset`	The input dataset.	required
`field`	`str`	A stateful field.	required
`metadata_fields`	`Optional[list[str]]`	Optional. The metadata fields to group by. If provided, it will take precedence over the schema metadata. Default is None.	`None`
`k_gram`	`Optional[int]`	Optional. The number of states to consider in each transition. Default is 2. If None, the full collapsed transitions for all states will be considered, regardless of the value of collapse.	`2`
`collapse`	`bool`	Whether to collapse repeated consecutive states into a single state. Default is False. This parameter is effective only when k_gram is not None.	`False`