Traveller

same here @mrwyattii

@mrwyattii Looking forward to your reply

> Contributor Thank you very much for your reply! I would like to ask again about the use of pipeline, that is traversing my own test set and inputting it...

> Deadlock can happen with both persistent and non-persistent modes. We don't check if there is enough memory to compute all the potential generated tokens for each request before placing...

@mrwyattii Can you please look at why this problem is occurring, I'm trying to test my own data in persistent mode, but it's going wrong in the warm up phase：...

@mrwyattii Your help is very much needed, using persistence has not been very smooth, why is it that when running on my own dataset, there is always a situation where...