hopwise.data.dataloader.general_dataloader¶

Classes¶

Module Contents¶

class hopwise.data.dataloader.general_dataloader.TrainDataLoader(config, dataset, sampler, shuffle=False)¶

Bases: hopwise.data.dataloader.abstract_dataloader.NegSampleDataLoader

TrainDataLoader is a dataloader for training. It can generate negative interaction when training_neg_sample_num is not zero. For the result of every batch, we permit that every positive interaction and its negative interaction must be in the same batch.

Parameters:

• config (Config) – The config of dataloader.

• dataset (Dataset) – The dataset of dataloader.

• sampler (Sampler) – The sampler of dataloader.

• shuffle (bool, optional) – Whether the dataloader will be shuffle after a round. Defaults to False.

logger¶

sample_size¶

_init_batch_size_and_step()¶

Initializing step and batch_size.

update_config(config)¶

Update configure of dataloader, such as batch_size, step etc.

Parameters:

config (Config) – The new config of dataloader.

collate_fn(index)¶

Collect the sampled index, and apply neg_sampling or other methods to get the final data.

class hopwise.data.dataloader.general_dataloader.NegSampleEvalDataLoader(config, dataset, sampler, shuffle=False)¶

Bases: hopwise.data.dataloader.abstract_dataloader.NegSampleDataLoader

NegSampleEvalDataLoader is a dataloader for neg-sampling evaluation. It is similar to TrainDataLoader which can generate negative items, and this dataloader also permits that all the interactions corresponding to each user are in the same batch and positive interactions are before negative interactions.

Parameters:

• config (Config) – The config of dataloader.

• dataset (Dataset) – The dataset of dataloader.

• sampler (Sampler) – The sampler of dataloader.

• shuffle (bool, optional) – Whether the dataloader will be shuffle after a round. Defaults to False.

logger¶

_init_batch_size_and_step()¶

Initializing step and batch_size.

update_config(config)¶

Update configure of dataloader, such as batch_size, step etc.

Parameters:

config (Config) – The new config of dataloader.

collate_fn(index)¶

Collect the sampled index, and apply neg_sampling or other methods to get the final data.

class hopwise.data.dataloader.general_dataloader.FullSortEvalDataLoader(config, dataset, sampler, shuffle=False)¶

Bases: hopwise.data.dataloader.abstract_dataloader.AbstractDataLoader

FullSortEvalDataLoader is a dataloader for full-sort evaluation. In order to speed up calculation, this dataloader would only return the data samples with positives, not negatives

Parameters:

• config (Config) – The config of dataloader.

• dataset (Dataset) – The dataset of dataloader.

• sampler (Sampler) – The sampler of dataloader.

• shuffle (bool, optional) – Whether the dataloader will be shuffle after a round. Defaults to False.

logger¶

check_sequential(config)¶

_build_positive_samples(dataset, sampler, feat, target_field, extra_fields=None)¶

_set_source_property(source, used_ids, positives)¶

_init_batch_size_and_step()¶

Initializing step and batch_size.

update_config(config)¶

Update configure of dataloader, such as batch_size, step etc.

Parameters:

config (Config) – The new config of dataloader.

_not_sequential_collate_fn(index, source_field)¶

collate_fn(index)¶

Collect the sampled index, and apply neg_sampling or other methods to get the final data.

class hopwise.data.dataloader.general_dataloader.FullSortRecEvalDataLoader(config, dataset, sampler, shuffle=False)¶

Bases: FullSortEvalDataLoader

FullSortRecEvalDataLoader is a dataloader for full-sort evaluation for the recommendation (Rec) task.

Parameters:

• config (Config) – The config of dataloader.

• dataset (Dataset) – The dataset of dataloader.

• sampler (Sampler) – The sampler of dataloader.

• shuffle (bool, optional) – Whether the dataloader will be shuffle after a round. Defaults to False.

uid_field¶

iid_field¶

_source_field¶

uid2items_num¶

uid2positive_item¶

uid2history_item¶

uid_list = []¶

user_df¶

class hopwise.data.dataloader.general_dataloader.FullSortLPEvalDataLoader(config, dataset, sampler, shuffle=False)¶

Bases: FullSortEvalDataLoader

FullSortLPEvalDataLoader is a dataloader for full-sort evaluation for the link prediction (LP) task.

Parameters:

• config (Config) – The config of dataloader.

• dataset (Dataset) – The dataset of dataloader.

• sampler (Sampler) – The sampler of dataloader.

• shuffle (bool, optional) – Whether the dataloader will be shuffle after a round. Defaults to False.

head_entity_field¶

relation_field¶

tail_entity_field¶

_source_field¶

head2tails_num¶

head2positive_tail¶

head2history_tail¶

head_list = []¶

kg_df¶