From a23f95e47b2bb974f3e099137cf95af62b7c2716 Mon Sep 17 00:00:00 2001 From: Myhs-phz Date: Mon, 7 Apr 2025 08:56:57 +0000 Subject: [PATCH] fix --- opencompass/summarizers/default.py | 17 ++++++++++++----- 1 file changed, 12 insertions(+), 5 deletions(-) diff --git a/opencompass/summarizers/default.py b/opencompass/summarizers/default.py index 88dd793b..e3765a42 100644 --- a/opencompass/summarizers/default.py +++ b/opencompass/summarizers/default.py @@ -111,8 +111,12 @@ class DefaultSummarizer: _dm = sorted(_dm, key=lambda i: METRIC_WHITELIST.index(i) if i in METRIC_WHITELIST else len(METRIC_WHITELIST)) if dataset_abbr in dataset_metrics: - assert tuple(dataset_metrics[dataset_abbr]) == tuple(_dm), \ - f'{dataset_abbr} has different metrics: {dataset_metrics[dataset_abbr]} vs {_dm}' + if len(_dm) == 1: + if _dm in dataset_metrics[dataset_abbr]: + dataset_metrics[dataset_abbr] = _dm + else: + assert tuple(dataset_metrics[dataset_abbr]) == tuple(_dm) or dataset_metrics[dataset_abbr][0] in _dm, \ + f'{dataset_abbr} has different metrics: {dataset_metrics[dataset_abbr]} vs {_dm}' else: dataset_metrics[dataset_abbr] = _dm parsed_results[model_abbr][dataset_abbr] = _rst @@ -197,9 +201,12 @@ class DefaultSummarizer: parsed_results[model_abbr][dataset_abbr][metric_default] eval_modes.append(dataset_eval_mode.get(dataset_abbr, 'unknown')) else: - scores.setdefault(metric, {})[dataset_abbr + '@' + metric] = \ - parsed_results[model_abbr][dataset_abbr][metric] - eval_modes.append(dataset_eval_mode.get(sg['subsets'][0], 'unknown')) + try: + scores.setdefault(metric, {})[dataset_abbr + '@' + metric] = \ + parsed_results[model_abbr][dataset_abbr][metric] + eval_modes.append(dataset_eval_mode.get(sg['subsets'][0], 'unknown')) + except: + continue result = {} for metric in scores: if default_metric == 'standard_deviation':