Error when I try to duplicate
Hi team,
Thanks all for your work !
I tried to duplicate for using my openAI API key but when building I got this error :
ArgillaCredentialsError: Argilla SDK error: ArgillaCredentialsError: Credentials
(api_key and/or api_url) are invalid
I used the key found on my Argilla setting space :
Can you help me on that issue ?
Thanks by advance.
Regards,
PEM
PS : Is the description of the dataset is just here for "imagine" the system prompt or is it here for another things ?
@BlackMinstrel your remark about the system prompts is correct.
w.r.t. the argilla errors
perhaps you are not passing the URL and KEY correctly.
If you're using Spaces, sometimes the Argilla UI is embedded into the Hub UI so the URL of the browser won't match the API URL. In these scenarios, you have several options:
On Argilla's Home page, click "Import from the SDK." The code snippet provided will contain your API URL and key.
Click on the three points menu at the top of the Space, select "Embed this Space", and open the direct URL.
Use this pattern: https://[your-owner-name]-[your_space_name].hf.space.
There is some more info here: https://docs.argilla.io/dev/getting_started/quickstart/#install-and-connect-the-python-sdk
Can you try to duplicate the space again with the new config?
Thanks for your answer, David.
I tried again with your advice, but I still got the same issue.
I think I did something wrong... I'll try again tomorrow and tel you here the result.
In the meantime, I’ll continue working with Python.
@BlackMinstrel would you be able to test if this works with the recently upgraded examples? https://github.com/argilla-io/synthetic-data-generator/tree/main/examples
@davidberenstein1957 Can you roughly tell me the steps to follow to properly conduct this test?
Hi @BlackMinstrel , you should be able to duplicate the space and set the environment variables outlined in the examples attached. You can for example follow the following example: https://github.com/argilla-io/synthetic-data-generator/blob/main/examples/hf-serverless-deployment.py.
@BlackMinstrel , has this happened after duplicating the space?
As with other spaces, requests can be queued due to usage by other users, hence the mention of a queue in the overview. Hosting your own version should resolve this.