Datasets:

Modalities:
Tabular
Text
Formats:
parquet
Size:
< 1K
ArXiv:
Libraries:
Datasets
pandas
bigcodebench-elo / README.md
terryyz's picture
Upload dataset
d26c9ca verified
|
raw
history blame
635 Bytes
---
dataset_info:
features:
- name: model
dtype: string
- name: lower
dtype: float64
- name: rating
dtype: float64
- name: upper
dtype: float64
- name: error_y
dtype: float64
- name: error_y_minus
dtype: float64
- name: rating_rounded
dtype: float64
splits:
- name: task_no_tie
num_bytes: 5127
num_examples: 72
- name: benchmark_tie
num_bytes: 5127
num_examples: 72
download_size: 16969
dataset_size: 10254
configs:
- config_name: default
data_files:
- split: task_no_tie
path: data/task_no_tie-*
- split: benchmark_tie
path: data/benchmark_tie-*
---