def transform_into_traffic_dataset(_, trip_dataset: TripDataFrame) -> TrafficDataFrame:
def max_traffic_load(trips):
interval_count = {
start_interval: 0 for start_interval in date_range(trips.name, periods=24, freq='h')
}
for interval in interval_count.keys():
upper_bound_interval = interval + timedelta(hours=1)
# Count number of bikes in transit during sample interval
interval_count[interval] = len(
trips[
(
( # Select trip if the trip started within the sample interval
(interval <= trips['start_time'])
& (trips['start_time'] < upper_bound_interval)
)
| ( # Select trip if the trip ended within the sample interval
(interval <= trips['end_time'])
& (trips['end_time'] < upper_bound_interval)
)
| ( # Select trip if the trip started AND ended outside of the interval
(trips['start_time'] < interval)
& (trips['end_time'] >= upper_bound_interval)
)
)
]
)
return max(interval_count.values())
counts = trip_dataset.groupby(['interval_date']).apply(max_traffic_load)
traffic_dataset = DataFrame(counts).reset_index()
traffic_dataset.columns = ['interval_date', 'peak_traffic_load']
return TrafficDataFrame(traffic_dataset)
def preprocess_trip_dataset(_, dataframe: DataFrame) -> TripDataFrame:
dataframe = dataframe[['bike_id', 'start_time', 'end_time']].dropna(how='all').reindex()
dataframe['bike_id'] = dataframe['bike_id'].astype('int64')
dataframe['start_time'] = to_datetime(dataframe['start_time'])
dataframe['end_time'] = to_datetime(dataframe['end_time'])
dataframe['interval_date'] = dataframe['start_time'].apply(lambda x: x.date())
return TripDataFrame(dataframe)