You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I'm just writing down these notes while I remember where I've go to 😄
Issues so far:
neural_lam makes direct calls to wandb which need to be replaced e.g. wandb.log(...) with pl.LightningModule.logger.log_metrics(...), pl.LightningModule.logger.log_image(...)
I have started on support for logging to mlflow on a branch: https://github.com/leifdenby/neural-lam/tree/feat/datastores-with-mlflow
I'm just writing down these notes while I remember where I've go to 😄
Issues so far:
neural_lam
makes direct calls towandb
which need to be replaced e.g.wandb.log(...)
withpl.LightningModule.logger.log_metrics(...)
,pl.LightningModule.logger.log_image(...)
WandbLogger.experiment.define_metric(...)
inneural_lam.utils.init_wandb_metrics(...)
yet: https://github.com/mllam/neural-lam/blob/main/neural_lam/utils.py#L268pytorch_lightning.MLFlowLogger
hasn't got.log_image(...)
implemented, although there appears to be support in the mlflow client to log images: https://learn.microsoft.com/en-us/azure/machine-learning/how-to-log-view-metrics?view=azureml-api-2&tabs=interactive#log-imagesThe text was updated successfully, but these errors were encountered: