metadata
dataset_info:
features:
- name: id
dtype: string
- name: ori_question
dtype: string
- name: ori_program
dtype: string
- name: question
dtype: string
- name: tests
sequence: string
- name: outputs
sequence: string
- name: scores
sequence: float64
- name: eval_results
list:
- name: code_error
dtype: string
- name: details
list:
- name: error_message
dtype: string
- name: pass
dtype: bool
- name: reason
dtype: string
- name: time_limit
dtype: float64
- name: pass_rate
dtype: float64
- name: status
dtype: string
- name: extracted_solutions
sequence: string
splits:
- name: train
num_bytes: 102496303
num_examples: 2000
download_size: 23099400
dataset_size: 102496303
configs:
- config_name: default
data_files:
- split: train
path: data/train-*