feature_table

feature_table

DescriptiveFeatures

DescriptiveFeatures(dataframe: DataFrame)

Compute descriptive statistics for a DataFrame, grouped by UUID.

Use compute to get descriptive statistics for each UUID in the DataFrame.

Attributes:

Name	Type	Description
data	DataFrame	DataFrame containing the data.

Initialize with the DataFrame to describe.

Parameters:

Name	Type	Description	Default
dataframe	DataFrame	DataFrame containing the data.	required

overall_stats

overall_stats(group: DataFrame) -> dict[str, int | float]

Compute and return overall statistics for the DataFrame group.

The returned dictionary contains:

• total_rows: Total number of rows in the group.
• total_time: Total time difference from max and min of 'systime' column.
• is_delta_sum: Sum of the 'is_delta' column.
• is_delta_avg: Mean of the 'is_delta' column.
• is_delta_std: Standard deviation of the 'is_delta' column.

The is_delta_* values are NaN when the group has no is_delta column (or no value in it).

Parameters:

Name	Type	Description	Default
group	DataFrame	DataFrame group (rows of a single UUID) to summarize.	required

Returns:

Name	Type	Description
dict	dict[str, int | float]	A dictionary with overall statistics.

compute_per_group

compute_per_group(
    group: DataFrame,
) -> dict[str, dict[str, Any]]

Compute and return statistics for each column in the DataFrame group.

Parameters:

Name	Type	Description	Default
group	DataFrame	DataFrame group (rows of a single UUID) to summarize.	required

Returns:

Name	Type	Description
dict	dict[str, dict[str, Any]]	A dictionary with overall statistics, and string, numeric, and boolean statistics per column. The long-format value columns are classified by name (value_bool boolean, value_double / value_integer numeric, value_string string), so every signal has the same keys; other columns by dtype. systime, uuid and is_delta are not described per column.

compute

compute(
    output_format: str = "dict",
) -> pd.DataFrame | dict[str, dict[str, dict[str, Any]]]

Compute and return descriptive statistics for each UUID in the DataFrame.

Parameters:

Name	Type	Description	Default
output_format	str	The desired output format ('dict' or 'dataframe'). Defaults to 'dict'.	'dict'

Returns:

Type	Description
DataFrame | dict[str, dict[str, dict[str, Any]]]	Union[DataFrame, dict]: A DataFrame or a nested dictionary with the UUID as the key and specific statistics related to that UUID's data type. An empty frame gives {} (or an empty DataFrame).

Raises:

Type	Description
ValueError	For an unknown output_format.
ColumnNotFoundError	If the frame has no uuid column.