Mario Souza
I'm really looking forward to having this feature pushed. @greghaskins @tjarratt Any chances for this to get merged?
@spiddy do you need help with that? I will be more than happy to keep the image up-to-date.
@sinzui How many events per second/minute are you processing? Is it possible to share the steps to reproduce it?
@pracucci I personally don't see a problem with it if 1) you have enough RAM and 2) RAM doesn't increase exponentionally as you start to process events. Why do you...
@iam404 Can you share what you expected and what is actually present in S3 after the data is uploaded?
If you use `json_lines` instead of `json`, the output would be as follows: ``` {foo: "bar"} {bar: "foo"} ``` Instead of: ``` [ {foo: "bar"}, {bar: "foo"} ] ```
Did not fix for me. I still got: ``` RuntimeError: CUDA error: the provided PTX was compiled with an unsupported toolchain. ``` I'm running llama-2-13b-chat-hf on 1xA100 40GB GPU.
My bad, I was trying to deploy to a host running nvidia driver 470.182.03 and CUDA 11.4. CUDA 11.7 (and 11.8) are not compatible with that nvidia driver version. Unfortunately...
@a9raag any chance to merge @leio10's pr to fix this issue?
@sam-h-bean have you managed to put a PR together for this?