I'm training a neural network built with pyTorch Lightning and I'm trying to have the HParams tab working in tensorboard.
Following the official outdated guide I initialize the logger with default_hp_metric=False:
logger = TensorBoardLogger("lightning_logs", name=modelName, default_hp_metric=False)
I set once the hyper parameters defining a custom metric:
logger.log_hyperparams(settings, metrics = {'val_maxroa': 0})
and I log to val_maxroa at the end of each validation step in the LightningModule:
def validation_step(self, batch, batch_idx):
inputs, target = batch
# ...Things...
self.log("val_maxroa", maxRoa)
On Tensorboard I can see the metric plotted in the "Time series" and "Scalars" tabs:
But in the HParams tab is not visible:
I checked also the official documentation but it's not very detailed for this aspect.
Do you have any hint?
