Logger

---

source

TENSORBOARDLOGGER

CLASS relax.logger.TensorboardLogger (log_dir, name, on_step=False)

Initialize self. See help(type(self)) for accurate signature.

logger = TensorboardLogger('log', name='debug')
logger.save_hyperparams({'lr': 0.01})
logger.log_dict({'train/train_loss_1': 0.1, 'epoch': 0})
logger.log_dict({'train/train_loss_1': 0.1})
logger.log_dict({'train/train_loss_1': 0.05})

logger.log_dict({'train/train_loss_1': 0.07, 'epoch': 1})
logger.log_dict({'train/train_loss_1': 0.05, })
logger.log_dict({'train/train_loss_1': 0.05, 'epoch': 2})

logger.close()