commit | 78657e1f8b2c97c3acc389e2b757c6ac8174388d | [log] [tgz] |
---|---|---|
author | AndrewZhaoLuo <andrew.zhao.luo@gmail.com> | Tue Apr 20 11:50:23 2021 -0700 |
committer | GitHub <noreply@github.com> | Tue Apr 20 12:50:23 2021 -0600 |
tree | 21eec40b93327edf573477eb02cb55e4c4fc0d8c | |
parent | bd2520fdae8181a56a7e4bef8b7f8408b6553393 [diff] |
[Relay] Add support for relay expressions as pad value for static pad (#7860) * add support for expr as inputs to pad * fix improper amount of args * add dynamic padding test * infer type better test * add comments to type relations * fix infer type layouts * proper return shape * proper shape infer type * make the tests pass by setting the conditions * make codegen reflect reality * make ternary operations more pythonic * proper infer layout * fold explicit padding * fix pattern matching in contrib * revert tests for contrib now that pattern matching works * revert import changes * add newline
Documentation | Contributors | Community | Release Notes
Apache TVM is a compiler stack for deep learning systems. It is designed to close the gap between the productivity-focused deep learning frameworks, and the performance- and efficiency-focused hardware backends. TVM works with deep learning frameworks to provide end to end compilation to different backends.
© Contributors Licensed under an Apache-2.0 license.
TVM adopts apache committer model, we aim to create an open source project that is maintained and owned by the community. Check out the Contributor Guide.
We learned a lot from the following projects when building TVM.