mathvista.yaml

dataset_path: AI4Math/MathVista
task: mathvista
test_split: testmini
output_type: "generate_until"
#process_docs: !function utils.process_docs
doc_to_image:
  - decoded_image
doc_to_text: "<image>{{query}}"
doc_to_target: answer
process_results: !function utils.process_results
generation_kwargs:
  until:
    - "<|endoftext|>"
  temperature: 0.0
  do_sample: false
  max_gen_toks: 1024
filter_list:
  - name: "extract_answer"
    filter:
      - function: "custom"
        filter_fn: !function utils.extract_all_answers
metric_list:
  - metric: acc
    aggregation: mean
    higher_is_better: true
metadata:
  version: 1.0
dataset_kwargs:
  trust_remote_code: true