Glad you found the culprit, hopefully you are okay now ๐ซ
Did your Caddy not automatically update it?
Why don't the first wish is "to give me unlimited wishes"? Unless the djinn has Terms& Condition ๐
You seems busy to manage all the servers ๐
By the way, is your email in GitHub still active? I would like to discuss something related to work or project later
A simple lightweight alternative API for language detection. It has compatible API route with LibreTranslate "/detect".
https://github.com/atrifat/language-detector-api
#OpenSource #Python
The API use same "langdetect" Python library but without including unused big translation models of LibreTranslate. Alternative faster "fast_langdetect" library can also be used by setting environment variable at startup.
#NostrFilterRelay admins/operators can also use this as their LibreTranslate alternative instance.
A simple lightweight alternative API for language detection. It has compatible API route with LibreTranslate "/detect".
https://github.com/atrifat/language-detector-api
#OpenSource #Python
Nice work ๐
How long the processing time of your fine tuning? Did you run it in your own GPU workstation?
I think it's not supported (yet) to write long form in Nostrudel. Nostrudel offer other app links to open/write long form.
Yeah, it is really expensive to fully rent GPU server ๐
Do you utilized fully of your workstation GPU memory or still have some left?
Thanks for replicate suggestion. Yes, the price quite cheap if the request process run shortly. But, it seems I need "private model" to be setup on their service.
Based on your experience, did they calculate purely based on how long the request to be processed? Let's say we send request every second but the request only need 100ms to be processed, they calculate only those 100ms right?
Congrats Kohei-san, wish you in a goodness ๐
nostr:npub1nxa4tywfz9nqp7z9zp7nr7d4nchhclsf58lcqt5y782rmf2hefjquaa6q8 Did one of your DVMs were hosted in 3rd party GPU private server or your own workstation?
Currently, I'm searching affordable GPU server besides of vast.ai and runpod.io . Do you have other recommendations?
In the past, I think we can select text which one that need to be copied. I think it's not available anymore, we can only copy whole text.
PSA:
Unfortunately, previous #NostrFilterRelay public demo relay ( wss://nostr-id-relay.hf.space ) was terminated by hosting provider due to highly excessive usage. We will move #Nostr relay into a new domain and new host provider in the next week(s).
Thank you and sorry for the inconveniences.
Oh, hopefully you can learn it better soon. I think you can also probably try to find Quran with Japanese translation, probably Muslim communities in Japan have that. ๐