-
Notifications
You must be signed in to change notification settings - Fork 6
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add documentation about how horovod works #35
Comments
Horovod accelerated PINNs: NeuralSolvers. Also has MirroredStrategy: https://github.com/tensordiffeq/TensorDiffEq Read: https://eng.uber.com/michelangelo-machine-learning-platform/ |
[Uber Open Summit 2018] Horovod: Deep Learning distribuido en 5 líneas de Python: https://www.youtube.com/watch?v=4y0TDK3KoCA&t=900s&ab_channel=UberEngineering |
This issue is stale because it has been open for 7 days with no activity. |
This issue was closed because it has been inactive for 7 days since being marked as stale. |
This issue was closed because it has been inactive for 7 days since being marked as stale. |
Here.
It could be interesting to add bibliography (from academia, and blogs (Medium, etc...).
Try to find the best articles to understand Horovod, how it works (technically), its structure.
I suggest to list the references, for example:
Documentation
Meet Horovod: Uber’s Open Source Distributed Deep Learning Framework for TensorFlow
Add a summary of what matters/you learnt or understood in this article
Distributed Deep Learning with Horovod
summary
and so on.
You can start with these references (and add your notes gradually):
https://towardsdatascience.com/distributed-deep-learning-with-horovod-2d1eea004cb2
https://arxiv.org/pdf/1706.02677.pdf
Horovod Article: https://arxiv.org/pdf/1802.05799.pdf
Uber presentation: https://developer.download.nvidia.com/video/gputechconf/gtc/2019/presentation/s9321-distributed-deep-learning-with-horovod.pdf
Add more references if you find more.
Important. Read: lululxvi/deepxde#39
The text was updated successfully, but these errors were encountered: