LocationClosestGranularity Class

Defines a closest location granularity.

_cord_count is the count of customer data after duplicates are dropped. If it's bigger than _cord_limit, the _lower_fuzzy_boundary is used to do the rough filtering, otherwise, _upper_fuzzy_boundary is used.

All possible locations will be ranked by spherical distance of two locations, the _closest_top_n are selected to do further joins.

Initialize with various configs.

cord_count is the count of customer_data after dropDuplicates(), if it's bigger than _cord_limit, we'll use _lower_fuzzy_boundary to do the rough filtering, otherwise, use _upper_fuzzy_boundary. all possible locations will be ranked by spherical distance of two locations, we'll select _closest_top_n to do further join.

Inheritance

LocationGranularity

LocationClosestGranularity

Constructor

LocationClosestGranularity(_cord_limit=5, _lower_fuzzy_boundary=2, _upper_fuzzy_boundary=5, _closest_top_n=1)

Parameters

Name	Description
_cord_limit	int The count of customer data after dropping duplicates. The default is 5. default value: 5
_lower_fuzzy_boundary	int Lower bound for filtering. The default is 2 default value: 2
_upper_fuzzy_boundary	int Upper bound for filtering. The default is 5 default value: 5
_closest_top_n	int How many of the top matches to consider closest. The default is 1. The bigger the value, the more time cost. default value: 1
_cord_limit Required	default is 5
_lower_fuzzy_boundary Required	default is 2
_upper_fuzzy_boundary Required	default is 5
_closest_top_n Required	default is 1, the bigger, the more time cost.

Methods

get_log_property

Get log property tuple, None if no property.

get_log_property

Get log property tuple, None if no property.

get_log_property()