§ u”j®ãó>—ddlmZmZddlmZGd„de¦«ZdS)é)ÚABCÚabstractmethod)ÚDictcóf—eZdZdZd„Zededefd„¦«Zed„¦«Zed d„¦«Z d S) ÚMetriczÙThe metric base class for computing metrics. The subclasses can either compute a single metric like 'accuracy', or compute the complex metrics for a specific task with or without other Metric subclasses. có—dS)N©)ÚselfÚargsÚkwargss úg/lsinfo/ai/hellotax_ai/data_center/backend/venv/lib/python3.11/site-packages/modelscope/metrics/base.pyÚ__init__zMetric.__init__ s€Ø ˆóÚoutputsÚinputscó—dS)a3 Append logits and labels within an eval loop. Will be called after every batch finished to gather the model predictions and the labels. Args: outputs: The model prediction outputs. inputs: The mini batch inputs from the dataloader. Returns: None Nr )r rrs r Úaddz Metric.adds €ð ˆrcó—dS)z²Evaluate the metrics after the eval finished. Will be called after the whole validation finished. Returns: The actual metric dict with standard names. Nr )r s r ÚevaluatezMetric.evaluates €ð ˆrÚothercó—dS)a* When using data parallel, the data required for different metric calculations are stored in their respective Metric classes, and we need to merge these data to uniformly calculate metric. Args: other: Another Metric instance. Returns: None Nr )r rs r Úmergez Metric.merge*s €ð ˆrN)rr) Ú__name__Ú __module__Ú __qualname__Ú__doc__rrrrrrr rr rrs–€€€€€ððð  ð ð ðð  ˜4ð  ¨ð  ð  ð  ñ„^ð  ðð ð ñ„^ð ðð  ð  ð  ñ„^ð  ð  ð  rrN)ÚabcrrÚtypingrrr rr úrscðà#Ð#Ð#Ð#Ð#Ð#Ð#Ð#ØÐÐÐÐÐð2 ð2 ð2 ð2 ð2 ˆSñ2 ô2 ð2 ð2 ð2 r