top of page
This site was designed with the
.com
website builder. Create your website today.Start Now
Search
  • Robert Kennedy
    • Jul 10, 2020
    • 1 min read

Week 5

This was another shortened week, leftover from the 4 day weekend. I worked on getting a mulit thor GNN setup to work on TensorFlow 2.x, but after quite a bit of struggle I switched back to TensorFlow 1.14. I think I plan on using 1.14 for the remainder of the summer. I was able to get a TF2 to work in a backwards compatibility mode, but I’m not sure this is what is best for the bundle. That will be up to the bundle owner. Plus, TF2 in a backwards compatibility mode might not have many of the 2.0 benefits you would want to use TF2 for in the first place. Larger changes to the core engine of the GNN bundle are needed to be fully TF2 compatible.

2 views0 comments

Recent Posts

See All

Week 11

This week all the code (major changes at least) is going to be unchanged. Other than cleaning it up for making a repo that others can use. No one wants to (or should have to) read poorly commented cod

Week 10

Nearing the home stretch. As part of trying to improve the efficiency of the GNN runtime across many GPUs, I started to experiment with using a different approach to using so many GPUs. A “normal” GNN

Week 9

It’s already August and there are only 3 more weeks until I run out of time with this internship! Time certainly flies. As per a recommendation from my mentor who knows way more about HPCC than I, I s

Post: Blog2_Post
bottom of page