src/backend/tasks/gsm8k/gsm8k-custom.yaml

group:
  - math_word_problems
task: gsm8k_custom
dataset_path: gsm8k
dataset_name: main
output_type: generate_until
training_split: train
fewshot_split: train
test_split: test
doc_to_text: "Question: {{question}}\nAnswer:"
doc_to_target: "{{answer}}" #" {{answer.split('### ')[-1].rstrip()}}"
metric_list:
  - metric: exact_match
    aggregation: mean
    higher_is_better: true
    ignore_case: true
    ignore_punctuation: false
    regexes_to_ignore:
      - ","
      - "\\$"
      - "(?s).*#### "
      - "\\.$"
generation_kwargs:
  until:
    - "<|eot_id|>"
  do_sample: false
  temperature: 0.0
  is_gsm8k: true
repeats: 1
num_fewshot: 5
filter_list:
  # - name: "strict-match"
  #   filter:
  #     - function: "regex"
  #       regex_pattern: "#### (\\-?[0-9\\.\\,]+)"
  #     - function: "take_first"
  - name: "flexible-extract"
    filter:
      - function: "regex"
        group_select: -1
        regex_pattern: "(-?[$0-9.,]{2,})|(-?[0-9]+)"
      - function: "take_first"
metadata:
  version: 3.0