1. Yes, because the codegen is now a part of TVM shared library. However, cmake
with ccache should save your time for unchanged files.
2. As the document indicates, you should apply a series of passes from
AnnotateTarget to PartitionGraph. After that, you could print the IR to see if
there's a
Thank you for answering. this helps quite a bit.
> I don’t understand this question, but a Relay program must have an output,
> and you cannot print message inside the Relay function.
Let's say for a a `relay.nn.conv2d` function, we produce a C function that
prints some information. The motiv
In the BYOC case you can control everything in your generated code. The runtime
only in charges of invoking the partitioned function on CPU by its name (i.e.,
symbol), so you could do anything you want. In short, it's doable to
generate/compile a C function with side effects and run it on a su
Hi,
I am trying to follow the instructions outlined here
https://tvm.apache.org/docs/how_to/tune_with_autotvm/tune_relay_arm.html to
autotune CNN on my android mobile phone (Samsung)
I have the following questions.
1. I am currently building tvm on Amazon AWS Ubuntu machine by "ssh"ing into
Right, that was my initial thoughts.
Although my intial question was: *can the generated code not bother about
filling the output tensor with some values?*
But I realise that the question is a stupid as `out` is already allocated,
whether its content is relevant or not shouldn't create any maj
