metadata
dataset_info:
features:
- name: premise
dtype: string
- name: hypothesis
dtype: string
- name: label
dtype: string
- name: premise_tptp
dtype: string
- name: hypothesis_tptp
dtype: string
- name: proof_inputs
sequence: string
- name: proof
dtype: string
- name: tc
dtype: int64
- name: 'n'
dtype: int64
- name: evidence_indices
sequence: int64
- name: counterfactuals
dtype: string
- name: evidences_necessity
dtype: bool
- name: mapping
dtype: string
splits:
- name: train
num_bytes: 96528440.52082026
num_examples: 3823
- name: validation
num_bytes: 12069211.239589872
num_examples: 478
- name: test
num_bytes: 12069211.239589872
num_examples: 478
download_size: 29723648
dataset_size: 120666863
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*