Spaces:
Running
on
Zero
new errors accessing api
A few days ago I started encountering errors when calling this space's API (predict/multimodal_understanding)
It used to work fine, but now I either get errors (including "The upstream Gradio app has raised an exception") or client.predict never responds.
I've tried troubleshooting on my end, so I'm wondering if it's a problem with the API.
Thanks
Update:
Most API calls fail, but some succeed; here trying every 2sec:
Loaded as API: https://deepseek-ai-janus-pro-7b.hf.space โ
Error on attempt 1 for 20240624T192941_DSCF0815.JPG:
Loaded as API: https://deepseek-ai-janus-pro-7b.hf.space โ
Error on attempt 2 for 20240624T192941_DSCF0815.JPG:
Loaded as API: https://deepseek-ai-janus-pro-7b.hf.space โ
Error on attempt 3 for 20240624T192941_DSCF0815.JPG:
Loaded as API: https://deepseek-ai-janus-pro-7b.hf.space โ
Error on attempt 4 for 20240624T192941_DSCF0815.JPG: The upstream Gradio app has raised an exception but has not enabled verbose error reporting. To enable, set show_error=True in launch().
Loaded as API: https://deepseek-ai-janus-pro-7b.hf.space โ
Error on attempt 5 for 20240624T192941_DSCF0815.JPG:
Loaded as API: https://deepseek-ai-janus-pro-7b.hf.space โ
Result for 20240624T192941_DSCF0815.JPG: No.
The errors raised can be:
- unspecified,
- The upstream Gradio app has raised an exception but has not enabled verbose error reporting. To enable, set show_error=True in launch().
- Error - can't start new thread
(I load to client before each call to avoid getting stuck on an unresponsive predict/multimodal_understanding call)