def __init__(
self,
start,
stop,
unit="km",
hours: Optional[Tuple[int, int]] = None,
table="all",
subscriber_identifier="msisdn",
ignore_nulls=True,
subscriber_subset=None,
):
self.unit = unit.lower()
if unit not in self.allowed_units:
raise ValueError(
f"Unrecognised unit {unit}, use one of {self.allowed_units}"
)
self.start = standardise_date(start)
self.stop = standardise_date(stop)
self.ul = SubscriberLocations(
self.start,
self.stop,
spatial_unit=make_spatial_unit("lon-lat"),
hours=hours,
table=table,
subscriber_subset=subscriber_subset,
subscriber_identifier=subscriber_identifier,
ignore_nulls=ignore_nulls,
)
super().__init__()
def __init__(
self,
start,
stop,
*,
hours: Optional[Tuple[int, int]] = None,
table="all",
subscriber_identifier="msisdn",
method="most-common",
subscriber_subset=None,
):
""""""
self.start = standardise_date(start)
self.stop = standardise_date(stop)
self.hours = hours
self.table = table
self.subscriber_identifier = subscriber_identifier
self.subscriber_tac = SubscriberTAC(
start,
stop,
hours=hours,
table=table,
subscriber_identifier=subscriber_identifier,
method=method,
subscriber_subset=subscriber_subset,
)
self.method = method
self.tacs = Table("infrastructure.tacs")
self.joined = self.subscriber_tac.join(self.tacs,
"tac",
"id",
how="left")
super().__init__()
def __init__(
self,
start: str,
stop: str,
*,
hours: Optional[Tuple[int, int]] = None,
table: Union[str, List[str]] = "all",
subscriber_identifier: str = "msisdn",
subscriber_subset: Optional[Query] = None,
):
self.start = standardise_date(start)
self.stop = standardise_date(stop)
self.hours = hours
self.tables = table
self.subscriber_identifier = subscriber_identifier
cols = [self.subscriber_identifier]
self.unioned = EventsTablesUnion(
self.start,
self.stop,
columns=cols,
tables=self.tables,
hours=hours,
subscriber_subset=subscriber_subset,
subscriber_identifier=self.subscriber_identifier,
)
super().__init__()
def __init__(
self,
start,
stop,
spatial_unit: AnySpatialUnit = make_spatial_unit("cell"),
hours: Optional[Tuple[int, int]] = None,
table="all",
subscriber_subset=None,
):
self.start = standardise_date(start)
self.stop = standardise_date(stop)
self.spatial_unit = spatial_unit
self.hours = hours
self.table = table
self.ul = UniqueLocations(
SubscriberLocations(
start=self.start,
stop=self.stop,
spatial_unit=self.spatial_unit,
hours=self.hours,
table=self.table,
subscriber_subset=subscriber_subset,
))
super().__init__()
def __init__(
self,
start,
stop,
statistic="avg",
*,
subscriber_identifier="msisdn",
hours: Optional[Tuple[int, int]] = None,
subscriber_subset=None,
):
self.start = standardise_date(start)
self.stop = standardise_date(stop)
self.subscriber_identifier = subscriber_identifier
self.hours = hours
self.statistic = statistic.lower()
self.tables = "events.topups"
if self.statistic not in valid_stats:
raise ValueError(
"{} is not a valid statistic. Use one of {}".format(
self.statistic, valid_stats))
column_list = [self.subscriber_identifier, "recharge_amount"]
self.unioned_query = EventsTablesUnion(
self.start,
self.stop,
tables=self.tables,
columns=column_list,
hours=hours,
subscriber_identifier=subscriber_identifier,
subscriber_subset=subscriber_subset,
)
super().__init__()
def __init__(
self,
start,
stop,
*,
hours="all",
table="all",
subscriber_identifier="msisdn",
subscriber_subset=None,
):
""""""
self.start = standardise_date(start)
self.stop = standardise_date(stop)
self.hours = hours
self.table = table
self.subscriber_identifier = subscriber_identifier
self.subscriber_tacs = SubscriberTACs(
start,
stop,
hours=hours,
table=table,
subscriber_identifier=subscriber_identifier,
subscriber_subset=subscriber_subset,
)
self.tacs = Table("infrastructure.tacs")
self.joined = self.subscriber_tacs.join(self.tacs,
"tac",
"id",
how="left")
super().__init__()
def __init__(
self,
start,
stop,
spatial_unit: Optional[AnySpatialUnit] = None,
hours: Optional[Tuple[int, int]] = None,
table="all",
subscriber_identifier="msisdn",
*,
ignore_nulls=True,
subscriber_subset=None,
):
""""""
self.start = standardise_date(start)
self.stop = standardise_date(stop)
if spatial_unit is None:
self.spatial_unit = make_spatial_unit("admin", level=3)
else:
self.spatial_unit = spatial_unit
self.hours = hours
self.table = table
self.subscriber_identifier = subscriber_identifier
self.subscriber_locs = SubscriberLocations(
start=self.start,
stop=self.stop,
spatial_unit=self.spatial_unit,
hours=self.hours,
table=self.table,
subscriber_identifier=self.subscriber_identifier,
ignore_nulls=ignore_nulls,
subscriber_subset=subscriber_subset,
)
super().__init__()
def __init__(
self,
start: str,
stop: str,
reference_location: BaseLocation,
statistic: str = "avg",
unit: str = "km",
hours: Union[str, Tuple[int, int]] = "all",
table: Union[str, List[str]] = "all",
subscriber_identifier: str = "msisdn",
ignore_nulls: bool = True,
return_subscribers_not_seen: bool = False,
subscriber_subset: Optional[Query] = None,
):
self.return_subscribers_not_seen = return_subscribers_not_seen
self.start = standardise_date(start)
self.stop = standardise_date(stop)
self.spatial_unit = reference_location.spatial_unit
subscriber_locations = SubscriberLocations(
self.start,
self.stop,
spatial_unit=self.spatial_unit,
hours=hours,
table=table,
subscriber_identifier=subscriber_identifier,
ignore_nulls=ignore_nulls,
subscriber_subset=subscriber_subset,
)
self.statistic = statistic.lower()
if self.statistic not in valid_stats:
raise ValueError(
"{} is not a valid statistic. Use one of {}".format(
self.statistic, valid_stats))
if not isinstance(reference_location, BaseLocation):
raise ValueError(
"Argument 'reference_location' should be an instance of BaseLocation class. "
f"Got: {type(reference_location)}")
else:
self.reference_location = reference_location
self.joined = reference_location.join(
other=subscriber_locations,
on_left=["subscriber"],
left_append="_from",
right_append="_to",
).join(
DistanceMatrix(spatial_unit=self.spatial_unit),
on_left=[
f"{col}_{direction}" for direction in ("from", "to")
for col in self.spatial_unit.location_id_columns
],
right_append="_dist",
how="left outer",
)
self.unit = unit
super().__init__()
def __init__(
self,
start,
stop,
*,
hours: Optional[Tuple[int, int]] = None,
table="all",
subscriber_subset=None,
subscriber_identifier="msisdn",
):
""""""
self.start = standardise_date(start)
self.stop = standardise_date(stop)
self.hours = hours
self.table = table
self.subscriber_identifier = subscriber_identifier
self.tbl = EventsTablesUnion(
start,
stop,
columns=[subscriber_identifier, "tac", "datetime"],
tables=table,
hours=hours,
subscriber_subset=subscriber_subset,
subscriber_identifier=self.subscriber_identifier,
)
super().__init__()
def __init__(
self,
start,
stop,
hours=(20, 4),
*,
subscriber_identifier="msisdn",
direction: Union[str, Direction] = Direction.BOTH,
subscriber_subset=None,
tables="all",
):
self.start = standardise_date(start)
self.stop = standardise_date(stop)
self.subscriber_identifier = subscriber_identifier
self.direction = Direction(direction)
self.hours = hours
self.tables = tables
column_list = [
self.subscriber_identifier,
"datetime",
*self.direction.required_columns,
]
self.unioned_query = EventsTablesUnion(
self.start,
self.stop,
tables=self.tables,
columns=column_list,
hours="all",
subscriber_identifier=subscriber_identifier,
subscriber_subset=subscriber_subset,
)
super().__init__()
def __init__(
self,
start,
stop,
*,
columns,
tables=None,
hours="all",
subscriber_subset=None,
subscriber_identifier="msisdn",
):
"""
"""
if isinstance(tables, str) and tables.lower() == "all":
logger.warn(
"EventsTablesUnion will soon stop accepting the argument tables='all'. Use tables=None instead."
)
tables = None
self.start = standardise_date(start)
self.stop = standardise_date(stop)
self.columns = columns
self.tables = self._parse_tables(tables)
if "*" in columns and len(self.tables) != 1:
raise ValueError(
"Must give named tables when combining multiple event type tables."
)
self.date_subsets = self._make_table_list(
hours=hours,
subscriber_subset=subscriber_subset,
subscriber_identifier=subscriber_identifier,
)
super().__init__()
def __init__(
self,
start,
stop,
*,
hours: Optional[Tuple[int, int]] = None,
table="all",
subscriber_subset=None,
subscriber_identifier="msisdn",
method="most-common",
):
""""""
if subscriber_identifier == "imei":
warnings.warn("IMEI has a one to one mapping to TAC number.")
self.start = standardise_date(start)
self.stop = standardise_date(stop)
self.hours = hours
self.table = table
self.subscriber_identifier = subscriber_identifier
self.subscriber_tacs = SubscriberTACs(
start,
stop,
hours=hours,
table=table,
subscriber_identifier=subscriber_identifier,
subscriber_subset=subscriber_subset,
)
self.method = method
if self.method not in ("most-common", "last"):
raise ValueError("{} is not a valid method.".format(method))
super().__init__()
def __init__(
self,
start,
stop,
*,
subscriber_identifier="msisdn",
statistic="sum",
spatial_unit: Optional[AnySpatialUnit] = None,
hours="all",
subscriber_subset=None,
):
self.start = standardise_date(start)
self.stop = standardise_date(stop)
self.subscriber_identifier = subscriber_identifier
if spatial_unit is None:
self.spatial_unit = make_spatial_unit("admin", level=3)
else:
self.spatial_unit = spatial_unit
self.statistic = statistic.lower()
if self.statistic not in valid_stats:
raise ValueError(
"{} is not a valid statistic. Use one of {}".format(
self.statistic, valid_stats
)
)
column_list = [
"id",
self.subscriber_identifier,
"msisdn_counterpart",
"outgoing",
"duration",
"location_id",
"datetime",
]
unioned_query = location_joined_query(
EventsTablesUnion(
self.start,
self.stop,
tables="events.calls",
columns=column_list,
hours=hours,
subscriber_subset=subscriber_subset,
subscriber_identifier=self.subscriber_identifier,
),
spatial_unit=self.spatial_unit,
time_col="datetime",
)
self.joined = unioned_query.subset("outgoing", "t").join(
unioned_query.subset("outgoing", "f"),
on_left="id",
on_right="id",
right_append="_counterpart",
how="left",
)
warnings.warn("This query is considerably slower than the other variants.")
super().__init__()
def __init__(
self,
start: str,
stop: str,
*,
spatial_unit: Optional[LonLatSpatialUnit] = None,
departure_rate: Union[pd.DataFrame, float] = 0.1,
hours: Union[str, Tuple[int, int]] = "all",
method: str = "last",
table: Union[str, List[str]] = "all",
subscriber_identifier: str = "msisdn",
subscriber_subset: Optional[Query] = None,
):
warnings.warn(
"The PopulationWeightedOpportunities model is currently **experimental**. "
+ "Please review Yan X-Y et al. " +
"(http://dx.doi.org/10.1098/rsif.2014.0834) " +
"before using this model in production.")
if isinstance(departure_rate, pd.DataFrame):
# Rename the columns to match what we'll join to
# sort the dataframe so we'll have a consistent md5
self.departure_rate = departure_rate.rename(
columns=lambda x: x if x == "rate" else f"{x}_from").apply(
lambda x: x.sort_values().values)
self.departure_rate = self.departure_rate.reindex(
columns=sorted(self.departure_rate.columns))
elif isinstance(departure_rate, float):
self.departure_rate = departure_rate
else:
raise TypeError(f"{departure_rate} must be a float or dataframe")
self.start = standardise_date(start)
self.stop = standardise_date(stop)
if spatial_unit is None:
self.spatial_unit = make_spatial_unit("versioned-site")
else:
self.spatial_unit = spatial_unit
self.distance_matrix = DistanceMatrix(spatial_unit=self.spatial_unit,
return_geometry=True)
self.population_object = ModalLocation(*[
daily_location(
d,
spatial_unit=self.spatial_unit,
hours=hours,
method=method,
table=table,
subscriber_identifier=subscriber_identifier,
ignore_nulls=True,
subscriber_subset=subscriber_subset,
) for d in list_of_dates(self.start, self.stop)
]).aggregate()
self.population_buffer_object = _PopulationBuffer(
population_object=self.population_object,
distance_matrix=self.distance_matrix,
)
def __init__(
self,
start=None,
stop=None,
*,
table="all",
total_by="day",
network_object: AnySpatialUnit = make_spatial_unit("cell"),
spatial_unit: Optional[AnySpatialUnit] = None,
hours: Optional[Tuple[int, int]] = None,
subscriber_subset=None,
subscriber_identifier="msisdn",
):
self.start = standardise_date(get_db().min_date(
table=table) if start is None else start)
self.stop = standardise_date(get_db().max_date(
table=table) if stop is None else stop)
self.table = table
if isinstance(self.table, str):
self.table = self.table.lower()
if self.table != "all" and not self.table.startswith("events"):
self.table = "events.{}".format(self.table)
network_object.verify_criterion("is_network_object")
self.network_object = network_object
if spatial_unit is None:
self.spatial_unit = make_spatial_unit("admin", level=0)
else:
self.spatial_unit = spatial_unit
# No sense in aggregating network object to network object
self.spatial_unit.verify_criterion("is_network_object", negate=True)
events = location_joined_query(
EventsTablesUnion(
self.start,
self.stop,
tables=self.table,
columns=["location_id", "datetime"],
hours=hours,
subscriber_subset=subscriber_subset,
subscriber_identifier=subscriber_identifier,
),
spatial_unit=self.network_object,
time_col="datetime",
)
self.joined = location_joined_query(events,
spatial_unit=self.spatial_unit,
time_col="datetime")
self.total_by = total_by.lower()
if self.total_by not in valid_periods:
raise ValueError("{} is not a valid total_by value.".format(
self.total_by))
super().__init__()
def __init__(self, start_date, end_date):
self.start_date = parse_datestring(start_date)
self.end_date = parse_datestring(end_date)
self.start_date_as_str = standardise_date(start_date)
self.end_date_as_str = standardise_date(end_date)
self.one_day_past_end_date = self.end_date + dt.timedelta(days=1)
self.one_day_past_end_date_as_str = standardise_date(
self.one_day_past_end_date)
def __init__(
self,
*,
subscriber_locations: SubscriberLocations,
reference_location: Union[BaseLocation, Tuple[float, float]] = (0, 0),
statistic: str = "avg",
time_bucket: str = "day",
):
subscriber_locations.spatial_unit.verify_criterion("has_geography")
subscriber_locations.spatial_unit.verify_criterion("has_lon_lat_columns")
self.spatial_unit = subscriber_locations.spatial_unit
if time_bucket.lower() in valid_time_buckets:
self.aggregate_by = time_bucket.lower()
else:
raise ValueError(
f"'{time_bucket}' is not a valid value for time_bucket. Use one of {valid_time_buckets}"
)
if statistic.lower() not in valid_stats:
raise ValueError(
f"'{statistic}' is not a valid statistic. Use one of {valid_stats}"
)
self.statistic = statistic.lower()
self.start = standardise_date(subscriber_locations.start)
self.stop = standardise_date(subscriber_locations.stop)
if isinstance(reference_location, tuple):
self.reference_location = reference_location
self.joined = subscriber_locations
elif isinstance(reference_location, BaseLocation):
if reference_location.spatial_unit != subscriber_locations.spatial_unit:
raise ValueError(
"reference_location must have the same spatial unit as subscriber_locations."
)
self.reference_location = reference_location
self.joined = reference_location.join(
other=subscriber_locations,
on_left=["subscriber"],
left_append="_from",
right_append="_to",
).join(
DistanceMatrix(spatial_unit=self.spatial_unit),
on_left=[
f"{col}_{direction}"
for direction in ("from", "to")
for col in self.spatial_unit.location_id_columns
],
right_append="_dist",
how="left outer",
)
else:
raise ValueError(
"Argument 'reference_location' should be an instance of BaseLocation class or a tuple of two floats. "
f"Got: {type(reference_location).__name__}"
)
super().__init__()
def __init__(
self,
start,
stop,
statistic="avg",
*,
hours="all",
tables="all",
direction: Union[str, Direction] = Direction.BOTH,
subscriber_subset=None,
exclude_self_calls=True,
):
self.tables = tables
self.start = standardise_date(start)
self.stop = standardise_date(stop)
self.hours = hours
self.direction = Direction(direction)
self.exclude_self_calls = exclude_self_calls
self.statistic = statistic.lower()
if self.statistic not in valid_stats:
raise ValueError(
"{} is not a valid statistic. Use one of {}".format(
self.statistic, valid_stats))
column_list = [
"msisdn", "msisdn_counterpart", "id", "location_id", "outgoing"
]
self.tables = tables
# EventsTablesUnion will only subset on the subscriber identifier,
# which means that we need to query for a unioned table twice. That has
# a considerable negative impact on execution time.
self.unioned_from_query = EventsTablesUnion(
self.start,
self.stop,
columns=column_list,
tables=self.tables,
subscriber_identifier="msisdn",
hours=hours,
subscriber_subset=subscriber_subset,
)
self.unioned_to_query = EventsTablesUnion(
self.start,
self.stop,
columns=column_list,
tables=self.tables,
subscriber_identifier="msisdn_counterpart",
hours=hours,
subscriber_subset=subscriber_subset,
)
self.distance_matrix = DistanceMatrix()
super().__init__()
def __init__(
self,
start,
stop,
proportion=0.8,
*,
direction: Union[str, Direction] = Direction.BOTH,
tables="all",
subscriber_identifier="msisdn",
hours: Optional[Tuple[int, int]] = None,
exclude_self_calls=False,
subscriber_subset=None,
):
self.start = standardise_date(start)
self.stop = standardise_date(stop)
self.hours = hours
self.direction = Direction(direction)
self.tables = tables
self.subscriber_identifier = subscriber_identifier
self.exclude_self_calls = exclude_self_calls
if 1 > proportion > 0:
self.proportion = proportion
else:
raise ValueError(
"{} is not a valid proportion.".format(proportion))
self.contact_balance = ContactBalance(
self.start,
self.stop,
hours=self.hours,
tables=self.tables,
subscriber_identifier=self.subscriber_identifier,
direction=self.direction,
exclude_self_calls=exclude_self_calls,
subscriber_subset=subscriber_subset,
)
self.subscriber_degree = SubscriberDegree(
self.start,
self.stop,
hours=self.hours,
tables=self.tables,
subscriber_identifier=self.subscriber_identifier,
direction=self.direction,
exclude_self_calls=self.exclude_self_calls,
subscriber_subset=subscriber_subset,
)
self._cols = ["subscriber", "pareto"]
super().__init__()
def __init__(
self,
start: str,
stop: str,
*,
table: Union[None, List[str]] = None,
spatial_unit: AnySpatialUnit = make_spatial_unit("cell"),
interval: str = "hour",
direction: Union[str, Direction] = Direction.BOTH,
hours: Optional[Tuple[int, int]] = None,
subscriber_subset=None,
subscriber_identifier="msisdn",
):
self.start = standardise_date(start)
self.stop = standardise_date(stop)
self.table = table
self.spatial_unit = spatial_unit
self.interval = interval
self.direction = Direction(direction)
if self.interval not in self.allowed_intervals:
raise ValueError("'Interval must be one of: {} got: {}".format(
self.allowed_intervals, self.interval))
self.time_cols = ["(datetime::date)::text AS date"]
if self.interval == "hour" or self.interval == "min":
self.time_cols.append("extract(hour FROM datetime) AS hour")
if self.interval == "min":
self.time_cols.append("extract(minute FROM datetime) AS min")
events_tables_union_cols = [
"location_id", "datetime", subscriber_identifier
]
# if we need to filter on outgoing/incoming calls, we will also fetch this
# column. Don't fetch it if it is not needed for both efficiency and the
# possibility that we might want to do pass another data type which does not
# have this information.
events_tables_union_cols += self.direction.required_columns
self.unioned = location_joined_query(
EventsTablesUnion(
self.start,
self.stop,
tables=self.table,
columns=events_tables_union_cols,
hours=hours,
subscriber_subset=subscriber_subset,
subscriber_identifier=subscriber_identifier,
),
spatial_unit=self.spatial_unit,
time_col="datetime",
)
super().__init__()
def __init__(
self,
start,
stop,
*,
subscriber_identifier="msisdn",
direction: Union[str, Direction] = Direction.OUT,
statistic="sum",
spatial_unit: Optional[AnySpatialUnit] = None,
hours="all",
subscriber_subset=None,
):
self.start = standardise_date(start)
self.stop = standardise_date(stop)
self.subscriber_identifier = subscriber_identifier
self.direction = Direction(direction)
if spatial_unit is None:
self.spatial_unit = make_spatial_unit("admin", level=3)
else:
self.spatial_unit = spatial_unit
self.statistic = statistic.lower()
if self.statistic not in valid_stats:
raise ValueError(
"{} is not a valid statistic. Use one of {}".format(
self.statistic, valid_stats
)
)
column_list = [
self.subscriber_identifier,
"msisdn_counterpart",
"duration",
"location_id",
"datetime",
*self.direction.required_columns,
]
self.unioned_query = location_joined_query(
EventsTablesUnion(
self.start,
self.stop,
tables="events.calls",
columns=column_list,
hours=hours,
subscriber_subset=subscriber_subset,
subscriber_identifier=self.subscriber_identifier,
),
spatial_unit=self.spatial_unit,
time_col="datetime",
)
super().__init__()
def __init__(
self,
start,
stop,
statistic="avg",
*,
spatial_unit: AnySpatialUnit = make_spatial_unit("cell"),
hours="all",
tables="all",
subscriber_identifier="msisdn",
direction: Union[str, Direction] = Direction.BOTH,
subscriber_subset=None,
):
self.start = standardise_date(start)
self.stop = standardise_date(stop)
self.spatial_unit = spatial_unit
self.hours = hours
self.tables = tables
self.subscriber_identifier = subscriber_identifier
self.direction = Direction(direction)
self.statistic = statistic
if self.statistic not in valid_stats:
raise ValueError(
"{} is not a valid statistic. Use one of {}".format(
self.statistic, valid_stats))
column_list = [
self.subscriber_identifier,
"location_id",
"datetime",
*self.direction.required_columns,
]
self.unioned_query = location_joined_query(
EventsTablesUnion(
self.start,
self.stop,
tables=self.tables,
columns=column_list,
hours=hours,
subscriber_identifier=subscriber_identifier,
subscriber_subset=subscriber_subset,
),
spatial_unit=self.spatial_unit,
time_col="datetime",
)
super().__init__()
def __init__(
self,
start: str,
total_periods: int,
period_length: int = 1,
period_unit: str = "days",
hours: Union[str, Tuple[int, int]] = "all",
table: Union[str, List[str]] = "all",
subscriber_identifier: str = "msisdn",
subscriber_subset: Optional[Query] = None,
):
self.start = standardise_date(start)
self.total_periods = total_periods
self.period_length = period_length
if period_unit not in self.allowed_units:
raise ValueError("`period_unit` must be one of {}".format(
self.allowed_units))
self.period_unit = period_unit
self.starts, self.stops = self._get_start_stops()
# For convenience also store when the whole thing ends
self.stop_date = time_period_add(
self.start, self.total_periods * self.period_length)
# This will be a long form table of unique subscribers in each time period
# i.e. a subscriber can appear more than once in this list, up to a maximum
# of the total time periods.
self.unique_subscribers_table = self._get_unioned_subscribers_list(
hours=hours,
table=table,
subscriber_identifier=subscriber_identifier,
subscriber_subset=subscriber_subset,
)
super().__init__()
def __init__(
self,
start,
stop,
numerator,
*,
numerator_direction: Union[str, Direction] = Direction.BOTH,
subscriber_identifier="msisdn",
direction: Union[str, Direction] = Direction.BOTH,
hours: Optional[Tuple[int, int]] = None,
subscriber_subset=None,
tables="all",
):
self.start = standardise_date(start)
self.stop = standardise_date(stop)
self.subscriber_identifier = subscriber_identifier
self.direction = Direction(direction)
self.numerator_direction = Direction(numerator_direction)
self.hours = hours
self.tables = tables
self.numerator = numerator if isinstance(numerator,
list) else [numerator]
self.numerator_query = EventCount(
self.start,
self.stop,
subscriber_identifier=self.subscriber_identifier,
direction=self.numerator_direction,
hours=self.hours,
subscriber_subset=subscriber_subset,
tables=self.numerator,
)
self.denominator_query = EventCount(
self.start,
self.stop,
subscriber_identifier=self.subscriber_identifier,
direction=self.direction,
hours=self.hours,
subscriber_subset=subscriber_subset,
tables=self.tables,
)
super().__init__()
def __init__(
self,
start,
stop,
characteristic,
hours="all",
table="all",
subscriber_identifier="msisdn",
method="most-common",
subscriber_subset=None,
):
"""
"""
self.start = standardise_date(start)
self.stop = standardise_date(stop)
self.hours = hours
self.table = table
self.subscriber_identifier = subscriber_identifier
self.characteristic = characteristic
if self.characteristic not in valid_characteristics:
raise ValueError("{} is not a valid characteristic.".format(characteristic))
if method == "most-common":
self.subscriber_handsets = SubscriberHandsets(
start,
stop,
hours=hours,
table=table,
subscriber_identifier=subscriber_identifier,
subscriber_subset=subscriber_subset,
)
else:
self.subscriber_handsets = SubscriberHandset(
start,
stop,
hours=hours,
table=table,
subscriber_identifier=subscriber_identifier,
subscriber_subset=subscriber_subset,
method=method,
)
self.method = method
super().__init__()
def __init__(
self,
start,
stop,
volume="total",
statistic="sum",
*,
subscriber_identifier="msisdn",
hours: Optional[Tuple[int, int]] = None,
subscriber_subset=None,
):
self.start = standardise_date(start)
self.stop = standardise_date(stop)
self.subscriber_identifier = subscriber_identifier
self.hours = hours
self.volume = volume
self.statistic = statistic.lower()
self.tables = "events.mds"
if self.statistic not in valid_stats:
raise ValueError(
"{} is not a valid statistic. Use one of {}".format(
self.statistic, valid_stats
)
)
if self.volume not in {"total", "upload", "download"}:
raise ValueError(f"{self.volume} is not a valid volume.")
column_list = [self.subscriber_identifier, f"volume_{self.volume}"]
self.unioned_query = EventsTablesUnion(
self.start,
self.stop,
tables=self.tables,
columns=column_list,
hours=hours,
subscriber_identifier=subscriber_identifier,
subscriber_subset=subscriber_subset,
)
super().__init__()
def __init__(
self,
start: str,
stop: str,
statistic: str = "avg",
*,
hours: Union[str, Tuple[int, int]] = "all",
tables: Union[str, List[str]] = "all",
subscriber_identifier: str = "msisdn",
subscriber_subset: Optional[Query] = None,
direction: Union[str, Direction] = Direction.OUT,
):
self.start = standardise_date(start)
self.stop = standardise_date(stop)
self.hours = hours
self.tables = tables
self.subscriber_identifier = subscriber_identifier
self.direction = Direction(direction)
column_list = [
self.subscriber_identifier,
"datetime",
*self.direction.required_columns,
]
self.statistic = statistic.lower()
if self.statistic not in valid_stats:
raise ValueError(
"{} is not a valid statistic. Use one of {}".format(
self.statistic, valid_stats))
self.unioned_query = EventsTablesUnion(
self.start,
self.stop,
tables=self.tables,
columns=column_list,
hours=self.hours,
subscriber_identifier=self.subscriber_identifier,
subscriber_subset=subscriber_subset,
)
super().__init__()
def __init__(
self,
start,
stop,
*,
location,
spatial_unit: AnySpatialUnit = make_spatial_unit("cell"),
hours="all",
table="all",
subscriber_identifier="msisdn",
ignore_nulls=True,
subscriber_subset=None,
):
"""
"""
if location == "any" and spatial_unit != make_spatial_unit("cell"):
raise ValueError(
"Invalid parameter combination: location='any' can only be used with cell spatial unit."
)
self.start = standardise_date(start)
self.stop = standardise_date(stop)
self.location = location
self.ul = SubscriberLocations(
self.start,
self.stop,
spatial_unit=spatial_unit,
hours=hours,
table=table,
subscriber_identifier=subscriber_identifier,
ignore_nulls=ignore_nulls,
subscriber_subset=subscriber_subset,
)
self.table = self.ul.table
self.subscriber_identifier = self.ul.subscriber_identifier
super().__init__()
def __init__(
self,
start,
stop,
*,
min_calls,
subscriber_identifier="msisdn",
direction: Union[str, Direction] = Direction.BOTH,
spatial_unit: Optional[AnySpatialUnit] = None,
hours: Optional[Tuple[int, int]] = None,
subscriber_subset=None,
):
from ...features import PerLocationSubscriberCallDurations
self.start = standardise_date(start)
self.stop = standardise_date(stop)
self.min_calls = min_calls
self.subscriber_identifier = subscriber_identifier
self.direction = Direction(direction)
if spatial_unit is None:
self.spatial_unit = make_spatial_unit("admin", level=3)
else:
self.spatial_unit = spatial_unit
self.pslds = PerLocationSubscriberCallDurations(
start=self.start,
stop=self.stop,
subscriber_identifier=self.subscriber_identifier,
direction=self.direction,
spatial_unit=self.spatial_unit,
statistic="count",
hours=hours,
subscriber_subset=subscriber_subset,
)
self.pslds_subset = self.pslds.numeric_subset("value",
low=self.min_calls,
high=inf)
super().__init__()
def __init__(
self,
start,
stop,
*,
hours: Optional[Tuple[int, int]] = None,
tables="all",
exclude_self_calls=True,
subscriber_subset=None,
):
self.tables = tables
self.start = standardise_date(start)
self.stop = standardise_date(stop)
self.hours = hours
self.exclude_self_calls = exclude_self_calls
self.tables = tables
self.contact_in_query = ContactBalance(
self.start,
self.stop,
hours=self.hours,
tables=self.tables,
subscriber_identifier="msisdn",
direction=Direction.IN,
exclude_self_calls=self.exclude_self_calls,
subscriber_subset=subscriber_subset,
)
self.contact_out_query = ContactBalance(
self.start,
self.stop,
hours=self.hours,
tables=self.tables,
subscriber_identifier="msisdn",
direction=Direction.OUT,
exclude_self_calls=self.exclude_self_calls,
subscriber_subset=subscriber_subset,
)
super().__init__()