metadata
dataset_info:
features:
- name: model_name
dtype: string
- name: benchmark
dtype: string
- name: accuracy
dtype: float64
- name: precision
dtype: float64
- name: recall
dtype: float64
- name: f1_score
dtype: float64
- name: pearson_correlation
dtype: 'null'
- name: non_parsed_rate
dtype: float64
splits:
- name: train
num_bytes: 626
num_examples: 8
download_size: 3926
dataset_size: 626
configs:
- config_name: default
data_files:
- split: train
path: data/train-*