Thanks for the RFC. Would you please file a formal RFC here:
https://github.com/apache/tvm-rfcs along with the upstream plan? For example,
you have some TODOs in this PR, so it would be good to have a formal RFC and
tracking issue for the progress.
--
Reply to this email directly or view it on
CC: @zxybazh @comaniac @merrymercy
--
Reply to this email directly or view it on GitHub:
https://github.com/apache/tvm/pull/9875#issuecomment-1007900534
You are receiving this because you are subscribed to this thread.
Message ID:
Thanks for the proposal! We are very interested in improving search algorithms
and cost model. I was very excited to read about FamilySeer a week ago.
In terms of the subgraph similarity, AFAIK @comaniac and @zxybazh have been
working independently on this topic to improve overall search time
### Motivation:
Auto-scheduler (Ansor) uses code sketch and optimization rules to generate a
large search space. The search space defined by Ansor has shown great
opportunities and therefore the search quality and the search efficiency are
determined by how we search the space.
Ansor utilize
We write an
Pre-RFC(https://discuss.tvm.apache.org/t/rfc-familyseer-a-new-search-method-for-auto-scheduler/11877
) and waiting for comments from reviewers.
You can view, comment on, or merge this pull request online at:
https://github.com/apache/tvm/pull/9875
-- Commit Summary --
* Family
How to specify the customized codegen when building the model?
should it be something like this?
```
relay.build(mod, newCodeGen, params=params)
```
where the `newCodeGen` is my customized codegen.
Or should it be something like this based on the cutlass example
```
tvm.get_global_func("relay.e
@denise-k @tqchen @jroesch @junrushao1994 @hogepodge @masahi @vinx13 @comaniac
@yzhliu @zhiics @icemelon @Hzfengsy @kparzysz-quic @mdw-octoml @mousius @u99127
@manupa-arm @leandron
--
Reply to this email directly or view it on GitHub:
https://github.com/apache/tvm-rfcs/pull/50#issuecomment-100
You can view, comment on, or merge this pull request online at:
https://github.com/apache/tvm-rfcs/pull/50
-- Commit Summary --
* Add Roadmap RFC
-- File Changes --
A rfcs/0050-roadmaps.md (216)
A rfcs/assets/0050/prior-art-1.jpeg (0)
A rfcs/assets/0050/prior-art-public-gh-roa
should we use remote.cuda() ?
Setup:
Remote machine: nvidia jetson agx xavier, linux-aarch64
Host: linux pc , x_86
I tried it gave following output
ctx = remote.cuda()
print(ctx.exist)
True
Is there any guide of using RPC to connect with gpu device
---
[Visit
Topic](https://di
