[bug] Tensorboard gives "No dashboards are active for the current data set." when passing minio path.

[pandapool]

Environment

• How do you deploy Kubeflow Pipelines (KFP)?
• KFP version:

  1.8.1

• KFP SDK version:

  1.8.19

  Steps to reproduce

  Create a function to check logging of scalars: def foo(runid) ->NamedTuple("EvaluationOutput", [('mlpipeline_ui_metadata', 'UI_metadata')]):

  import subprocess subprocess.run(["pip","install","tensorboardX==2.5"]) from tensorboardX import SummaryWriter from minio import Minio

  log_dir = " /tensorboard_logs/" writer = SummaryWriter(log_dir)

  for i in range(20):

  Log loss to TensorBoard

  acc = (i+2)**2/(i+1)**2 + i-1 
  writer.add_scalar("Accuracy",acc, i+1)

  writer.flush()

  from pipelines.utils.minio_utils import upload_local_directory_to_minio
  upload_local_directory_to_minio(client, bucket_name, " /tensorboard_logs/","foo_logs")

  metadata = { 'outputs': [{ 'type': 'tensorboard', 'source': f'http://:/minio/datasets/foo_logs', }] }

  out_tuple = namedtuple("EvaluationOutput", ["mlpipeline_ui_metadata"]) return out_tuple(json.dumps(metadata))

Expected result

Tensorboard should show a graph for the scalars written to events file. The issue is not with the events file as it works for a locally deployed tensorboard server.

Materials and reference

Labels

---

Impacted by this bug? Give it a 👍.

[github-actions[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[github-actions[bot]]

This issue has been automatically closed because it has not had recent activity. Please comment "/reopen" to reopen it.