I’ve been using Hugging Face to train and upload my models. Recently, I encountered an issue where I cannot access the training metrics through TensorBoard for the models I’ve uploaded. Here is the code snippet I used for uploading my models to Hugging Face:
from huggingface_hub import HfApi
api = HfApi()
api.upload_folder(
folder_path="/path/to/local/model",
repo_id="username/model",
repo_type="model",
)
Previously, I was able to view the TensorBoard for my uploaded models without any issues. However, now I’m facing a problem where TensorBoard shows “No dashboards are active for the current data set” with the following probable causes listed:
You haven’t written any data to your event files.
TensorBoard can’t find your event files.
I have not made any changes to my code, so I don’t believe the issue lies there. The last reload was on Feb 22, 2024, 11:36:02 AM, and the log directory is /data/daniel246-drug-class-2023-12-5-59e7b/data
You can view the error message here: Daniel246/drug_class_2023_12_5_s_640_1 · Training metrics.
I’m seeking advice on why my uploaded models are no longer working with training metrics and what steps I can take to resolve this issue. Any help or guidance would be greatly appreciated. Thank you.