How can I get the final value of all the metrics that have been logged

[pamparana34]

At the moment, I am calling trainer.fit(model=model, datamodule=datamodule) and the training works fine.
However, what I want to do is return the final value of all the metrics that have been logged during the training.

In my case, the training and validation steps looks as follows:

def validation_epoch_end(self, outputs: EPOCH_OUTPUT) -> None:
        loss = torch.stack(outputs).mean()
        self.log("val/loss", loss)

def training_epoch_end(self, outputs: EPOCH_OUTPUT) -> None:
        loss = torch.stack(outputs).mean()
        self.log("train/loss", loss)

Basically, at the end of the training, I would like to query the final values of train/loss and val/loss and basically every other metrics that have been logged.

[rohitgr7]

trainer.callback_metrics <- to access all the metrics
trainer.logged_metrics <- to access only the logged metrics