You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently the server will just request an answer from the next service if it does not get an answer from the first source. This makes the service very slow. Speed is an important thing to keep users onboard. A fast user experience is very very important.
EXPECTED:
Therefore a good alternative is: If the system is not able to get an answer from a source it could just answer something like: “I tried to get an answer for your questions by searching [source here], but I was not able to get a good response. Please ask me something else.”
The text was updated successfully, but these errors were encountered:
We already have timeouts for each API call (by default), but we don't have a timeout for the whole thinking process of susi server. That means, it can in principle continue for a long long time. (Max ideas is I think 100).
Maybe we need a total time limit until an answer needs to be found, and if no answer is found by that time a null answer is returned.
Great, fine with me. It shouldn't be too difficult to add a simple timeout feature to the one call. But it might be better to implement that in susi_python directly, at least that is what I suggest
Currently the server will just request an answer from the next service if it does not get an answer from the first source. This makes the service very slow. Speed is an important thing to keep users onboard. A fast user experience is very very important.
EXPECTED:
Therefore a good alternative is: If the system is not able to get an answer from a source it could just answer something like: “I tried to get an answer for your questions by searching [source here], but I was not able to get a good response. Please ask me something else.”
The text was updated successfully, but these errors were encountered: