README.md

metadata

dataset_info:
  features:
    - name: problem_id
      dtype: int64
    - name: stem
      dtype: large_string
    - name: reference_code
      dtype: large_string
    - name: reference_path
      dtype: large_string
    - name: input_tensor_spec_path
      dtype: large_string
    - name: world_size
      dtype: int64
    - name: default_m
      dtype: int64
    - name: default_n
      dtype: int64
    - name: default_dtype
      dtype: large_string
    - name: default_trials
      dtype: int64
  splits:
    - name: train
      num_bytes: 281750
      num_examples: 87
  download_size: 87364
  dataset_size: 281750
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*