# Dataset information: the entry point for dataset configuration, including file locations for each modality.
dataset_id: StoreSales
dataset_name: Store_Sales
domain: retail_supply_chain

sample_defination:
  sample_type: store-family inventory demand node
  sample_id_graph: episode_id # Corresponding graph node identifier column.
  sample_id_time_series: episode_id # Corresponding time series identifier column.
  description: each sample corresponds to one (store_nbr, family) node; episode_id is int 0..N-1 aligned across modalities

modalities:
# Available feature modalities.
  meta_features: # Static metadata, relative to input_multimodal/ under the protocol root.
    required: true # meta_features is required.
    file_name: meta_features.csv # Combining this with dataset.py's _meta_candidates yields input_multimodal/meta_features.csv.

  node_level_task_graphs:
    available: true # Whether a graph for node-level tasks is available.
    graph_ids:
      - store_family_relation_graph

  time_series:
    available: true # Whether time series data is available.
    time_series_ids:
      - sales
      - onpromotion
      - dcoilwtico
      - is_holiday

tasks: # Task construction.
# Supported task types.
  - store_type_classification
  - sales_flow_forecasting

# Data source information.
license:
  data_license: Kaggle competition terms (source-dependent)
  redistribution_allowed: true

source:
  name: Corporacion Favorita Grocery Sales Forecasting
  url: https://www.kaggle.com/competitions/favorita-grocery-sales-forecasting