Enrico Shippole

> Hello! Thank you for the response. I was attempting to use serialize on the Tensor Dataset in order to fix one of the errors thrown. It is possible that...

Hi @ultmaster, I do not believe that I need serialization after adding nni.trace to a few different functions and classes while testing today. The errors seem to be thrown due...

Hello @ultmaster , Thank you for the input! I will make the recommended changes as you had listed as I continue to work on this. After implementing everything, hopefully correctly...

@scarlett2018 Hi Scarlett, I have made some progress but have yet to fully resolve the issue. I still am having a reoccurring problem with running out of memory during the...

Hi @albertvillanova , Here is the environment information: ``` - `datasets` version: 2.3.2 - Platform: Linux-5.4.0-122-generic-x86_64-with-glibc2.27 - Python version: 3.9.12 - PyArrow version: 7.0.0 - Pandas version: 1.4.2 ``` Thanks,...

Hi @loubnabnl, Yes, the issue is solved in the discussion thread. I will close this issue. Thank you again for all of your help. Enrico

> code segment below will report error as titled, under multi gpu training > > ``` > # rotary embeddings > positions = self.get_rotary_embedding(n, device) > q, k = map(lambda...

Hi @lvwerra , I greatly appreciate you taking the time to answer my questions. I found the CodeParrot blog very insightful for causal lanugauge modeling. This is one of the...

Hi @ver217 , Thank you for confirming that the softmax when initialized as a certain type will not be overridden by the ColossalAI engine. I wanted to ensure that when...

@lucidrains Thank you for the additional info and for confirming that it is expected to be slower. And thanks to @ahennequ as well! I will run all of the benchmarks...