3

I am trying to store MLflow artifacts on a remote server running MLflow. The server I am accessing from and server running MLflow are both VMs on google cloud. I can see the matrices in the MLflow server but not the artifacts.

I tried the flollowing methods but nonoe of them is working:

  • mlflow server --backend-store-uri /mnt/persistent-disk --default-artifact-root /tmp/ --host=0.0.0.0
  • mlflow server --backend-store-uri /mnt/persistent-disk --default-artifact-root /path/to/folder/with/mlrun --host=0.0.0.0

I also gave rwx permissions to the path but still getting the same error :

PermissionError: [Errno 13] Permission denied: '/home/user/folder'
Mehdi
  • 1,260
  • 2
  • 16
  • 36
bazinga
  • 2,120
  • 4
  • 21
  • 35

1 Answers1

0

Can you please advise on the solution you found for this?

requesting since i am finding the artifacts are stored into the mlruns in cwd in the local server itself if i use mlflow.pyfunc.log_model instead of mlflow.sklearn.log_model(lr, "model")

i have created https://github.com/mlflow/mlflow/issues/1815

cyril_thank
  • 9
  • 1