dataset_path: PLM-Team/Spatial-Visualization-Benchmark

generation_kwargs:
  max_new_tokens: 8192
  temperature: 0.0
  top_p: 1.0
  num_beams: 1
  do_sample: false

output_type: generate_until
doc_to_visual: !function utils.spatialviz_doc_to_visual
doc_to_text: !function utils.spatialviz_doc_to_text
doc_to_target: utils.spatialviz_doc_to_target
process_results: !function utils.spatialviz_process_results

metric_list:
  - metric: spatialviz_score
    aggregation: !function utils.spatialviz_aggregate_results
    higher_is_better: true

dataset_kwargs:
  token: True
  cache_dir: SpatialViz
  force_download: true

# metadata:
#   strategy: CoT # ['Direct', 'CoT']
#   use_lmms_judge: False