Hi @mbrookhart, thanks.
I got it working by replacing the last line with
func = relay.Function(analysis.free_vars(new_body), new_body)
Not sure, why the one that I originally used not working. But this solves the
problem. Thanks.
[quote="mbrookhart, post:5, topic:9254"]
I think the issue is
I am currently trying to run TVM Hexagon on an Android device (Snapdragon).
Just to get started, I wish to offload a simple matrix multiplication on the
aDSP with TVM.
So far my understanding from reading the page "Introducing Hexagon Backend", I
need to build the following:
1) IDL Libraries
@dmitriy-arm Thanks for the reply :+1:
---
[Visit
Topic](https://discuss.tvm.apache.org/t/model-graph-size-doubling-after-partitioning-for-arm-compute-library/9131/7)
to respond.
You are receiving this because you enabled mailing list mode.
To unsubscribe from these emails, [click
here]
Hi @mbrookhart,
I have made some attempts as you suggested (doing reverse topological order and
getting the part after IF as you suggested). Here is my understanding of your
suggestions which I also agree with. Please let me know if I got them wrong.
Here are the steps that I assumed that I s
You only need to load one of the libraries, as libtvm already contains the
runtime component
---
[Visit
Topic](https://discuss.tvm.apache.org/t/check-failed-override-global-packedfunc-tvm-set-device-is-already-registered-failed/3111/4)
to respond.
You are receiving this because you enabl
Sorry for being non-verbosive. Yes there is an issue in current implementation
of Graph runtime led to local and external parameters being serialized twice
(in graph runtime module and then in metadata module). There is proposed fix
for this https://github.com/apache/tvm/pull/7564
Be aware th
I think @dmitriy-arm has been looking into this.
---
[Visit
Topic](https://discuss.tvm.apache.org/t/model-graph-size-doubling-after-partitioning-for-arm-compute-library/9131/4)
to respond.
You are receiving this because you enabled mailing list mode.
To unsubscribe from these emails, [cl
