libcity.data.dataset.dataset_subclass.pbs_trajectory_dataset

class libcity.data.dataset.dataset_subclass.pbs_trajectory_dataset.PBSTrajectoryDataset(config)

Bases: libcity.data.dataset.abstract_dataset.AbstractDataset

popularity based negative sampling weighted random sampling based on np.random.choice

cutter_filter()

切割后的轨迹存储格式: (dict)

{

uid: [

[

checkin_record, checkin_record, …

], [

checkin_record, checkin_record, …

}

divide_data()

Returns

train_data (list) eval_data (list) test_data (list)

encode_traj(data)

encode the cut trajectory

Parameters

data (dict) –

the key is uid, the value is the uid’s trajectories. For example: {

uid: [

trajectory1, trajectory2

]

} trajectory1 = [

dyna_id, dyna_id, …..

]

Returns

For example:

{

data_feature: {…}, pad_item: {…}, encoded_data: {uid: encoded_trajectories}

}

Return type

dict

get_data()

轨迹比较特殊，原子文件中存储的并不是轨迹而是一个一个点，因此需要先对轨迹进行切割

get_data_feature()

返回一个 dict，包含数据集的相关特征

Returns

包含数据集的相关特征的字典

Return type

dict

get_encoder()