Datasets:
metadata
language:
- en
license: mit
size_categories:
- 1K<n<10K
task_categories:
- time-series-forecasting
pretty_name: Next purchase day
configs:
- config_name: default
data_files:
- split: test
path: data/test-*
- split: train
path: data/train-*
- split: validation
path: data/validation-*
dataset_info:
features:
- name: start
dtype: timestamp[us]
- name: target
sequence: float32
- name: item_id
dtype: string
- name: feat_static_cat
sequence: uint64
- name: feat_dynamic_real
sequence:
sequence: float32
splits:
- name: test
num_bytes: 12143
num_examples: 34
- name: train
num_bytes: 11055
num_examples: 34
- name: validation
num_bytes: 11599
num_examples: 34
download_size: 29957
dataset_size: 34797