Hi there, after running relay.exp targetting at both llvm and cuda, I met a crash. Below is my script:
import tvm
from tvm import relay
var_0 = relay.var("var_0", dtype = "int64", shape = (9, 10)) # shape=(9, 10)
var_1 = relay.exp(var_0) # shape=(9, 10)
tuple = relay.Tuple([var_1,])
F = relay.Function([var_0], tuple)
mod = tvm.IRModule()
mod['main'] = F
mod = relay.transform.InferType()(mod)
print(mod.astext(show_meta_data=False))
graph, lib, params = relay.build(mod, target='cuda')
Here is the crash message:
And if the target is llvm, then the message is
Is it because my incorrect use of type? If it’s the case, why the use of int64 on exp is forbidden?