metadata
dataset_info:
features:
- name: sample_id
dtype: string
- name: product_name
dtype: string
- name: item_name
dtype: string
- name: item_description
dtype: string
- name: unit
dtype: string
- name: value
dtype: float32
- name: image_link
dtype: image
- name: price
dtype: float32
splits:
- name: train
num_bytes: 19937441361
num_examples: 70000
- name: validation
num_bytes: 272461347
num_examples: 1000
- name: test
num_bytes: 1140618086.672
num_examples: 3999
download_size: 17673521816
dataset_size: 21350520794.672
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*