Cannot find config for target=llvm -keys=cpu

Questions
#1

Hi there,

I’m running the x86 tutorial on a GCP VM instance, but got this warning below

Cannot find config for target=llvm -keys=cpu, workload=(‘dense_nopack.x86’, (‘TENSOR’, (1, 512), ‘float32’), (‘TENSOR’, (1000, 512), ‘float32’), None, ‘float32’). A fallback configuration is used, which may bring great performance regression.

The tuning still seems done but not sure if it optimizes the execution. Any idea? Thanks!

#2

@comaniac Hmm, not sure if you’re not the right person for this question, but I’d really appreciate it if you could provide some hints or pointers to others.

#3

Because the tutorial only tunes conv2d tasks:

tasks = autotvm.task.extract_from_program(
    mod["main"], target=target, params=params, ops=(relay.op.get("nn.conv2d"),)
)

Although I don’t think missing one dense schedule will hurt the performance a lot, you can modify the code to the following to let AutoTVM tune the dense op as well:

tasks = autotvm.task.extract_from_program(
    mod["main"], target=target, params=params)
)

When no ops is given, AutoTVM extracts all tasks by default.