-
Notifications
You must be signed in to change notification settings - Fork 28
about winograd transform matrix #6
Comments
No. |
Thanks for explanation.
I get the IR, but do not understand how TVM could remove all zero in this select statement. Thanks |
You should lower the whole kernel transform |
I write lower code like following
but I got following IR:
Could you help to check why my lower command is not right? But the generated cuda code already removed zero in G and s[G].compute_inline is necessary. Thanks |
I modify my lower code like following
I could get the right IR. Thanks |
hi, @merrymercy
in conv2d.py, winograd algorithm do G and B like normal matrix multiplication, this will not reduce multiplication by plus/minus. Is this understood correct?
Thanks
The text was updated successfully, but these errors were encountered: