Skip to content
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

Reduction Op Perf Model #1329

Open
ipotkonjak-tt opened this issue Nov 19, 2024 · 0 comments
Open

Reduction Op Perf Model #1329

ipotkonjak-tt opened this issue Nov 19, 2024 · 0 comments

Comments

@ipotkonjak-tt
Copy link

Goal

We aim to provide a model of the reduction op performance based on the possible combinations of input parameters. Because of the complexity of the op in question we will focus on modelling a simple control flow path (that excludes other op calls within reduction). First goal is to provide modelling PoC.

We will attempt to model the perf with simpler functions, like Linear Regression, and move on to more complex ones if there is a need. Training data will be split into logical segments and for each of them a different model will be created. For each of the models accuracy error will be provided, to give Optimizer information about the quality of the estimate.

Additional information and results will be provided in the comment section and sub-issues.
Op Perf Model - Parent Issue

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

2 participants