Reading Bert model using ClassificationModel is failing with utf-8' codec can't decode byte 0x80 in position 0: invalid start byte

Describe the bug A clear and concise description of what the bug is. Please specify the class causing the issue.

So I saved the classification model as a pickle file and I am trying to read it back it’s failing with the following error

‘utf-8’ codec can’t decode byte 0x80 in position 0: invalid start byte this is because the it was read using r mode in model_args.py line number 96 if read it using rb this should fix the problem

To Reproduce Steps to reproduce the behavior: Save a classification model and try read it back

Expected behavior A clear and concise description of what you expected to happen.

Screenshots If applicable, add screenshots to help explain your problem.

Desktop (please complete the following information):

• OS databrics cluster

Additional context Add any other context about the problem here.

[ThilinaRajapakse]

The saved model consists of all the files in the output directory. As the documentation says, it should be the path to the directory itself and not a particular file.

The example from the docs.

You don’t need to manually save the model. It gets saved during training according to the parameters you set (e.g. save_steps, save_model_every_epoch).

If your model files are in /dbfs/FileStore/tables/, then you would load them like so.

model = ClassificationModel("roberta", "/dbfs/FileStore/tables/")

[stale[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.