After the update, the based on my notes no longer works
Opened this issue ยท 19 comments
Can you send me a screenshot of the developer console logs when this happens?
That should help clarify what is causing the issue.
Thanks for the report and your help in figuring this out ๐ด
I have the same problem ... :-(
@brianpetro a mi tampoco me funciona con open ai. cualquiera de los modelos yo indexe todo el vaul ya el tema es el bug a veces no aparecen las conexiones y la busqueda no funciona en nada
@fabioscarsi @oscarvdrgrgrgr Can you send me a screenshot of any errors in the developer console logs when this happens? ๐ด
I'm wrong because it doesn't let you index with the open ai large model either. I have an API key along with tokens but it doesn't work with what I indexed a few versions ago.
@oscarvdrgrgrgr that 429 error suggests you need to add credits to your OpenAI account to use their embedding model. Occasionally things will change and that can require re-indexing. I'm not sure which version you're migrating from, but that could be why you need to re-index ๐ด
@liuling73 that shouldn't happen every open. Are their errors in the developer console logs when you open? ๐ด
@liuling73 have you tried with the "legacy transformers" setting turned off? ๐ด
Problem solved, thank you
a mi tambien se me soluciono si era la clave api el tema es que como no me aparecia ningun letrero indicando eso pense que no era.
@oscarvdrgrgrgr good to know, thanks for the follow up ๐ด
@brianpetro have similar error as OP and maybe as Oscar, however, not sure.
I do not follow how Oscar figured it out.
I am at at the latest public release of Smart connections 2.3.45
"Just says unknown error"
I have tried to:
re-index
uninstall Smart connections
remove smart.env
folder
tried different API keys on different Google accounts (I am using Gemini)
tried different models, I was using Gemini 1.5 Pro and 2.0 the other day without problems.
FYI
I can use Gemini 1.5 Pro and 2.0 with the Text Generator plugin, so accessing the API has no issue
Best regards,
Olivmamman
Problem solved, thank you
@liuling73 How did you solve this problem? I met the same problem too. It'll will make Smart Connections embbedding from beginning whenever restarting Obsidian. If I change embedding model from Jina-v2-base-zh-8k to other, it'll be normal. But I really need Jina-v2-base-zh-8k.
@liuling73 You disable "Embed Blocks", I don't, that's difference. I don't think this is a good solution.
@palwoo have you tried deleting the .smart-env/multi
folder to ensure a complete re-embedding? ๐ด