lmms_eval/tasks/vqav2/vqav2_val.yaml (10 lines of code) (raw):
task: "vqav2_val"
include: _default_template_vqav2_yaml
test_split: validation
metric_list:
- metric: exact_match
aggregation: mean
higher_is_better: true
ignore_case: true
ignore_punctuation: true
process_results: !function utils.vqav2_process_results_val