Oh, I see, thanks for your kind reply.
---
[Visit
Topic](https://discuss.tvm.apache.org/t/strassen-algorithm-for-dense/2661/17)
to respond.
You are receiving this because you enabled mailing list mode.
To unsubscribe from these emails, [click
here](https://discuss.tvm.apache.org/email/u
If you want to measure it more robust, you should run it more times and
calculate its average time. For example you could run 1000 times.
---
[Visit
Topic](https://discuss.tvm.apache.org/t/strassen-algorithm-for-dense/2661/16)
to respond.
You are receiving this because you enabled mailin
Thank you for your reply.
Regarding time-consuming fluctuations, I didn't make it clear.
After autotvm tune is completed, I picked the best record for time-consuming
testing, and its time-consuming fluctuates significantly.I calculate the time
difference between the start and the end to get t
On your case, current code is will call 4 cores (id 0 ~ 3). So parallel brings
you better performance.
About time consuming functions, Do you use auto tvm? If you use auto tvm, the
default cpu TVM uses is big core (that is index 7). If you decide to use 4
little cores, you should make auto tv
Thank you very much for your reply.
As I said before, I refer to this tutorial to deploy tvm:
https://tvm.apache.org/docs/deploy/cpp_deploy.html. I export tvm.build function
as a library first, then load and call the function in C++.
According to your suggestion, I set the cpu affinity in thi
I don't think u should set `TVM_NUM_THREADS` on arm because of arm's BIG LITTLE
architecture. I think you should call `runtime.config_thread_pool` to complete
the core binding work. Another thing is we shouldn't make tvm worker thread run
different frequency cpus (aka, one worker thread is in
Thank you very much for your reply.
Ansor looks great, I am very interested, I will try it.If there is any
progress, I will be happy to discuss further with you.
---
[Visit
Topic](https://discuss.tvm.apache.org/t/strassen-algorithm-for-dense/2661/11)
to respond.
You are receiving this b
Thank you very much for your reply.
The hardware I use is AArch64 CPU with 8 cores. I refer to this tutorial to
deploy tvm：https://tvm.apache.org/docs/deploy/cpp_deploy.html.The c++ thread
that load and use tvm library is bound to 3 intermediate frequency cpus, and
TVM_NUM_THREADS is set to 1
@jcf94 has explained very well for strassen algorithm. The link you posted is I
wrote. However, we should notice that my post is not to show the best
performance TVM could achieve, just show how easy TVM could a reasonable
performance (beyond numpy).
If we still want to improve performance,
Hi, it's nice to see strassen has attracted attention again. I would like to
know which hardware have you used and how many cores have you used?
Actually, it's easy to implement strassen in TVM, and I have tested this
algorithm with two different implementations.
TE version:
```python
def str
@FrozenGene
Thank you for your reply. I refer to this tutorial for my experiments:
https://zhuanlan.zhihu.com/p/75203171, and this tutorial seems to be written by
you. I tried [1024, 1024] * [1024, 1024], [2048, 2048] * [2048, 2048], [256,
256] * [256, 256], [128, 128] * [128, 128], etc, and
The performance can not beyond dense would have many reasons, but I think
strassen algorithm is not one key part. @jcf94 has done some experiment on this.
---
[Visit
Topic](https://discuss.tvm.apache.org/t/strassen-algorithm-for-dense/2661/6) to
respond.
You are receiving this because yo
I did some tests, and the result is that tvm cannot exceed MNN.
---
[Visit
Topic](https://discuss.tvm.apache.org/t/strassen-algorithm-for-dense/2661/5) to
respond.
You are receiving this because you enabled mailing list mode.
To unsubscribe from these emails, [click
here](https://discus
@FrozenGene，do you have any updates about this topic？
---
[Visit
Topic](https://discuss.tvm.apache.org/t/strassen-algorithm-for-dense/2661/4) to
respond.
You are receiving this because you enabled mailing list mode.
To unsubscribe from these emails, [click
here](https://discuss.tvm.apac
14 matches
Mail list logo
