MCPcopy Create free account
hub / github.com/apple/ml-pointersect / train_step

Method train_step

cdslib/core/script/base_train.py:1095–1105  ·  view source on GitHub ↗

One training step. Return a dictionary that will be passed to logging.

(
            self,
            epoch: int,
            bidx: int,
            batch: T.Any,
    )

Source from the content-addressed store, hash-verified

1093 @customer
1094 @abstractmethod
1095 def train_step(
1096 self,
1097 epoch: int,
1098 bidx: int,
1099 batch: T.Any,
1100 ) -> T.Dict[str, T.Any]:
1101 """One training step.
1102
1103 Return a dictionary that will be passed to logging.
1104 """
1105 raise NotImplementedError
1106
1107 @customer
1108 def visualize_train_step(

Callers

nothing calls this directly

Calls

no outgoing calls

Tested by

no test coverage detected