NEED HELP: uploading stuck

checkpoints directory : ./xxx/chkpt
Epoch: 1 0.006 MB of 0.028 MB uploaded
Steps : 0, XXX Loss Total : 68.359, XXX Error: 0.219, XXX. Error : 0.007, time since last 1 iters : 1.691
wandb: / 0.006 MB of 0.028 MB uploaded

It happened many times when I ran the code. It seems the uploading is frozen (the signal after wandb: / - \ - / (continue)… never finishes uploading for quite long time) and never goes to the next step.

Anyone can help me? Thank you so much in advance.

Hi @rhalona

Good day and thank you for reaching out to us. We’re sorry to hear that you are experiencing issues when uploading files.

May I know what is your current SDK version? You can get this by running wandb --version.

Also, to further investigate this, could you please share a code snippet so we could take a look at how the files are being uploaded.

Thanks,
Paulo

Hi @paulo-sabile

Thank you for the responses.

When I ran
wandb --version
displayed
wandb, version 0.17.7

What kind of snippet you want?

Hi @rhalona,

Good day! Could you share how you are uploading the artifacts? Are these from your local files or from an external source e.g. GCS or S3 bucket?

A code snippet on how the artifacts are uploaded should help. Additionally, did it ever upload completely or it posted errors?

Thanks,
Paulo

Hi @paulo-sabile,

Thank you for your response.

I uploaded the local files, and after restarting my PC, it works now.

I’m not sure what caused the issue, but it seems to be resolved.

Thank you again for your continued support.

Thank you for your update @rhalona ! I am happy to hear that it is now working for you. I will now mark this request as solved. Please feel free to reach out to us again if you need assistance!

Best Regards,
Paulo