A collection of research ready datasets for sequential recommendation
Sequential Recommendation Datasets
Provide a tool for helping dealing with some common sequential recommendation datasets
Install this tool
pip install -U srdatasets —-user
Run the command below to download datasets. As some datasets are not directly accessible, you'll be warned to download them manually and place them somewhere it tells you.
srdatasets download --dataset=[dataset_name]
To get a view of downloaded and processed status of all datasets, run
The generic processing command is
srdatasets process --dataset=[dataset_name] [--options]
Two dataset splitting methods are provided: user-based and time-based. User-based means that splitting is executed on every user behavior sequence given the ratio of validation set and test set, while time-based means that splitting is based on the date of user behaviors. After splitting some dataset, two processed datasets are generated, one for development, which uses the validation set as the test set, the other for test, which contains the full training set.
--split-by User or time (default: user) --test-split Proportion of test set to full dataset (default: 0.2) --dev-split Proportion of validation set to full training set (default: 0.1)
NOTE: time-based splitting need you to manually input days at console by tipping you total days of that dataset, since you may not know.
Task related options
For short term recommnedation task, you use previous
input-len items to predict next
target-len items. To make user interests more focused, user behavior sequences can also be cut into multiple sessions if
session-interval is given. If the number of previous items is smaller than
input-len, 0 is padded to the left.
For long and short term recommendation task, you use
pre-sessions previous sessions and current session to predict
target-len items. The target items are picked randomly or lastly from current session. So the length of current session is
target-len while the length of any previous session is
max-session-len. If any previous session or current session is shorter than the preset length, 0 is padded to the left.
--task Short or long-short (default: short) --input-len Number of previous items (default: 5) --target-len Number of target items (default: 1) --pre-sessions Number of previous sessions (default: 10) --pick-targets Randomly or lastly pick items from current session (default: random) --session-interval Session splitting interval (minutes) (default: 0) --min-session-len Sessions less than this in length will be dropped (default: 2) --max-session-len Sessions greater than this in length will be cut (default: 20)
--min-freq-item Items less than this in frequency will be dropped (default: 5) --min-freq-user Users less than this in frequency will be dropped (default: 5) --no-augment Do not use data augmentation (default: False) --remove-duplicates Remove duplicated items in user sequence or user session (if splitted) (default: False)
Dataset related options
--rating-threshold Interactions with rating less than this will be dropped (Amazon, Movielens, Yelp) (default: 4) --item-type Recommend artists or songs (Lastfm) (default: song)
By using different options, a dataset will have many processed versions. You can run the command below to get configurations and statistics of all processed versions of some dataset. The
config id shown in output is a required argument of
srdatasets info --dataset=[dataset_name]
DataLoader is a built-in class that makes loading processed datasets easy. Practically, once initialized a dataloder by passing the dataset name, processed version (config id), batch_size and a flag to load training data or test data, you can then loop it to get batch data. Considering that some models use rank-based learning, negative sampling is intergrated into DataLoader. The negatives are sampled from all items except items in current data according to popularity. By default it (
negatives_per_target) is turned off. Also, the time of user behaviors is sometimes an important feature, you can include it into batch data by setting
include_timestmap to True.
dataset_name: dataset name (case insensitive)
config_id: configuration id
batch_size: batch size (default: 1)
train: load training dataset (default: True)
development: load the dataset aiming for development (default: False)
negatives_per_target: number of negative samples per target (default: 0)
include_timestamp: add timestamps to batch data (default: False)
drop_last: drop last incomplete batch (default: False)
from srdatasets.dataloader import DataLoader trainloader = DataLoader("amazon-books", "c1574673118829", batch_size=32, train=True, negatives_per_target=5, include_timestamp=True) testloader = DataLoader("amazon-books", "c1574673118829", batch_size=32, train=False, include_timestamp=True)
For pytorch users, there is a wrapper implementation of
torch.utils.data.DataLoader, you can then set keyword arguments like
pin_memory to speed up loading data
from srdatasets.dataloader_pytorch import DataLoader trainloader = DataLoader("amazon-books", "c1574673118829", batch_size=32, train=True, negatives_per_target=5, include_timestamp=True, num_workers=8, pin_memory=True) testloader = DataLoader("amazon-books", "c1574673118829", batch_size=32, train=False, include_timestamp=True, num_workers=8, pin_memory=True)
For short term recommendation task
for epoch in range(10): # Train for users, input_items, target_items, input_item_timestamps, target_item_timestamps, negative_samples in trainloader: # Shape # users: (batch_size,) # input_items: (batch_size, input_len) # target_items: (batch_size, target_len) # input_item_timestamps: (batch_size, input_len) # target_item_timestamps: (batch_size, target_len) # negative_samples: (batch_size, target_len, negatives_per_target) # # DataType # numpy.ndarray or torch.LongTensor pass # Test for users, input_items, target_items, input_item_timestamps, target_item_timestamps in testloader: pass
For long and short term recommendation task
for epoch in range(10): # Train for users, pre_sessions_items, cur_session_items, target_items, pre_sessions_item_timestamps, cur_session_item_timestamps, target_item_timestamps, negative_samples in trainloader: # Shape # users: (batch_size,) # pre_sessions_items: (batch_size, pre_sessions * max_session_len) # cur_session_items: (batch_size, max_session_len - target_len) # target_items: (batch_size, target_len) # pre_sessions_item_timestamps: (batch_size, pre_sessions * max_session_len) # cur_session_item_timestamps: (batch_size, max_session_len - target_len) # target_item_timestamps: (batch_size, target_len) # negative_samples: (batch_size, target_len, negatives_per_target) # # DataType # numpy.ndarray or torch.LongTensor pass # Test for users, pre_sessions_items, cur_session_items, target_items, pre_sessions_item_timestamps, cur_session_item_timestamps, target_item_timestamps in testloader: pass
This repo does not host or distribute any of the datasets, it is your responsibility to determine whether you have permission to use the dataset under the dataset's license.
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
|Filename, size||File type||Python version||Upload date||Hashes|
|Filename, size srdatasets-0.1.4-py3-none-any.whl (30.0 kB)||File type Wheel||Python version py3||Upload date||Hashes View|
|Filename, size srdatasets-0.1.4.tar.gz (24.5 kB)||File type Source||Python version None||Upload date||Hashes View|
Hashes for srdatasets-0.1.4-py3-none-any.whl