-
Could you add more servers in FedAvg for faster training speed?
As the BytePS does.
-
TL;DR: MMTk core should be more friendly for VMs where each Mutator does not correspond to one language-level thread. To do this, we must reconsider the relation between **mutators**, **threads** and…
-
Something wrong about the dimension in deepspeed's auto-tp.py:
-
The swarming code is tightly coupled to the OPF but could be very useful elsewhere. It should be a library that has two main operations for a set of permute objects - ask for new parameter set, and gi…
-
First of all, I'd like to acknowledge the time and effort so many of you have put into this idea of Gonimo! Thank you for the idea and the realisation thereof.
In a failed attempt of getting a USB …
-
ZIO? Akka?
I've heard that Scala prefers Monads / etc for the problems that coroutines solve. Need to do more research into options to speed up the concurrency.
-
Hello AICB Team,
Thanks for your outstanding work on this open-source project—it's a fantastic resource for the community.
I am currently exploring the use of AICB to generate communication work…
-
## ❓ Questions and Help
I recived error when try create sqmd mesh on kaggle notebook when flow [Huggingface optimum-tpu](https://github.com/huggingface/optimum-tpu/blob/695ee84d657d9ed2761fcf481685af…
-
I'm looking into the problem of scaling SourceBrowser right now. Currently, running our entire codebase through the indexer takes ~28 hours on a pretty decent machine. I see two options:
1) Federat…
-
First of all, thanks for this great tool for parsing PDFs. I am facing issues when extracting text from two column text pages in PDF (research paper). In such cases, the start of the left column text …