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
We've confirmed in #16 that the maximum queue size has the potential of affecting overall performance. In fact, the number of items in the queue can eventually be used by the system to dynamically adjust step placement for handling an unstable input stream (a high number of items in the queue means that the system is having a hard time following the input stream). We should collect data on the number of items in the queue for our experiments to get more insights on how to use the information for dynamic step placement.
We could either create a separate log file for this information, or somehow embed it within the inference timing logs.
The text was updated successfully, but these errors were encountered:
We've confirmed in #16 that the maximum queue size has the potential of affecting overall performance. In fact, the number of items in the queue can eventually be used by the system to dynamically adjust step placement for handling an unstable input stream (a high number of items in the queue means that the system is having a hard time following the input stream). We should collect data on the number of items in the queue for our experiments to get more insights on how to use the information for dynamic step placement.
We could either create a separate log file for this information, or somehow embed it within the inference timing logs.
The text was updated successfully, but these errors were encountered: