metadata
dataset_info:
features:
- name: user_id
dtype: string
- name: item_asins
list: string
- name: item_asin
dtype: string
- name: seq_input_ids
list: int64
- name: seq_labels
dtype: int64
- name: history_item_title
list: string
- name: item_title
dtype: string
- name: history_rating
list: float64
- name: rating
dtype: float64
- name: history_timestamp
list: int64
- name: timestamp
dtype: int64
- name: interaction_id
dtype: int64
- name: prompt
dtype: string
- name: user_gen_input_ids
list: int32
- name: user_gen_attention_mask
list: int8
- name: profile
list: string
- name: completion_0
dtype: string
- name: user_input_ids
list: int32
- name: user_attention_mask
list: int8
- name: user_embedding
list: float64
- name: scores
list: float64
splits:
- name: train
num_bytes: 7051048158
num_examples: 68252
- name: valid
num_bytes: 885858851
num_examples: 8531
- name: test
num_bytes: 887005115
num_examples: 8532
download_size: 1173955204
dataset_size: 8823912124
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: valid
path: data/valid-*
- split: test
path: data/test-*