Giskard org
edited Mar 1

Fixed feedback from https://www.notion.so/giskard/3rd-QnA-Session-by-5aa63cfad688409db407e8d9026a30a1

No need to keep two links to the same HF token page. I would remove the text You can find it here

Model Wrap Advanced Config.

It would be more user-intuitive, if we swap the info and issue columns, because, for example, the info columns are on the left in the header, but in the table they appear on the right. It is a bit strange.

For the user it will be strange to have the tabular task, while the space is called 🐢Giskard Evaluator - Text Classification, as well as the Text Classification dashboard name. Either we make the name of the space more generic, or let’s remove temporary tabular task, until we support it.

Scanner Advanced Config

Do not like the phrase Giskard Bot Evaluation Guide: . Not intuitive to me, because It implies, that I need to do something to evaluate the model, but I already did.

  • Remove the : ending. We never use it for the section names.
  • Remove the leading whitespace.
  • “Your have successfully submitted a Giskard evaluation. Further details are available in the Logs tab. You can find your report will be posted to your model's community discussion.
    Your job id is: 57ec4f4f-adc1-4113-89ce-638d7d318957

from another tab phrase is unclear. Specify exactly that this tab is “Text Classification”. Also, it would be cool to have it as a hyper-link to the dedicated tab.

In the queue.

When I replace the model, the dataset selection is not reset.

We do not support an evaluation without Inference API , right ?

ZeroCommand changed pull request status to open
inoki-giskard changed pull request status to merged

Sign up or log in to comment