So in essence:
Normally a complete TE graph will be lowered to TIR representation. Your
assumptions is that doing small changes in part of the TE graph should not
propagate throughout all the TIR AST and you want to somehow "cache" the part
of the TIR which is independent of changes of this su
Hi Experts,
@tqchen, @kazum
I tried to auto tune the sample Resnet model on iOS metal target , I went
through the
(https://discuss.tvm.apache.org/t/auto-tvm-how-to-auto-tune-the-model-on-ios-device/7681).
while trying to tune the model every time I see **"Current/Best: 0.00/ 0.00
GFLOPS**
Essentially I'm trying to 'partition' a TE (that is, the graph of
tensors/ComputeOps). The reason I want to do this is because the graph is very
large (actually it's a whole network lowered to TE) and I want to try some
alternative scheduling options on only a small subgraph at a time. I can
I don't think I completely follow.
You have some chain of te stages. You want to pick a subset of these (i.e. the
subgraph) and only (?) lower this subgraph?
What happens to the rest of the stages?
Wouldn't tensorize be a way to technically handle a subset of the te stages
"differently" (i.e
Is there a way in Python that I can create a 'subgraph' from a Tensor
Expression? In particular, I have a large TE graph containing many operators
and would like to lower only a small subgraph from within it. I'd expect all
the inputs to the subgraph to be replaced with equivalent placeholder
I wanted to ask if someone has successfully trained a model with TVM? I found
some old
[discussions](https://discuss.tvm.apache.org/t/train-a-model-with-the-tvm/1892)
about training models, but unfortunately no answers. If someone is up to date
and can point me in the right direction I would
