diff --git a/TC-Eval/inference/aggregate_results.py b/TC-Eval/inference/aggregate_results.py index 7f29b0c..b9543c5 100644 --- a/TC-Eval/inference/aggregate_results.py +++ b/TC-Eval/inference/aggregate_results.py @@ -30,7 +30,7 @@ def _create_placeholder_eval_result_dict() -> Dict[str, Dict[str, Any]]: pdict['FGC'] = {str(k): {"response": _resp_status} for k, v in data.items()} # TTQA - data = json.load(open(f'{_CUR_DIR}/../data/TTQA/TTQA_mc_2.0.0.json')) + data = json.load(open(f'{_CUR_DIR}/../data/TTQA/TTQA_mc.json')) pdict['TTQA'] = {str(k): {"response": _resp_status} for k, v in data.items()} # XSum_TC @@ -141,4 +141,4 @@ def aggregate_inference_results_for_evaluation(self, inference_results_dir): result_dir = args.result_dir model_name = args.model_name agg = ResultAggregator(model_name) - agg.aggregate_inference_results_for_evaluation(result_dir) \ No newline at end of file + agg.aggregate_inference_results_for_evaluation(result_dir) diff --git a/TC-Eval/inference/scenarios.py b/TC-Eval/inference/scenarios.py index 2eedcd9..6bdfaeb 100644 --- a/TC-Eval/inference/scenarios.py +++ b/TC-Eval/inference/scenarios.py @@ -55,7 +55,7 @@ def __getitem__(self, idx) -> Dict[str, any]: class TTQA(Dataset): - def __init__(self, data_path: str = f"{_CUR_DIR}/../data/TTQA/TTQA_mc_2.0.0.json", **kwargs): + def __init__(self, data_path: str = f"{_CUR_DIR}/../data/TTQA/TTQA_mc.json", **kwargs): raw_data = json.load(open(data_path, "r", encoding="utf-8")) self._js_ds = [dict(id=str(i), **obj) for i, obj in raw_data.items()]