An internal invariant was violated during the execution of TVM

I want autotune a net and get best graph, but I can’t compile. I compiled successfully other net with same code.

Extract tasks… [07:46:13] /tmp/build/80754af9/libmxnet_1564766659613/work/src/nnvm/legacy_json_util.cc:209: Loading symbol saved by previous version v1.2.0. Attempting to upgrade… [07:46:13] /tmp/build/80754af9/libmxnet_1564766659613/work/src/nnvm/legacy_json_util.cc:217: Symbol successfully upgraded! Tuning… Compile… Traceback (most recent call last): File “auto_tune1.py”, line 149, in tune_and_evaluate(tuning_option)
File “auto_tune1.py”, line 125, in tune_and_evaluate mod, target=target, params=params) File “/home/kk/tvm/python/tvm/relay/build_module.py”, line 269, in build graph_json, mod, params = bld_mod.build(mod, target, target_host, params) File “/home/kk/tvm/python/tvm/relay/build_module.py”, line 132, in build self._build(mod, target, target_host) File "/home/kk/tvm/python/tvm/_ffi/ctypes/packed_func.py", line 237, in call raise get_last_ffi_error() tvm.ffi.base.TVMError: Traceback (most recent call last): [bt] (8) /home/kk/tvm/build/libtvm.so(tvm::relay::ExprFunctor<tvm::RelayExpr (tvm::RelayExpr const&)>::InitVTable()::{lambda(tvm::runtime::ObjectRef const&, tvm::relay::ExprFunctor<tvm::RelayExpr (tvm::RelayExpr const&)>)#6}::_FUN(tvm::runtime::ObjectRef const&, tvm::relay::ExprFunctor<tvm::RelayExpr (tvm::RelayExpr const&)>)+0x2c) [0x7f4cfc9a4fec] [bt] (7) /home/kk/tvm/build/libtvm.so(tvm::relay::MixedModeMutator::VisitExpr(tvm::relay::CallNode const*)+0x9e) [0x7f4cfc91cfce] [bt] (6) /home/kk/tvm/build/libtvm.so(tvm::relay::ForwardRewriter::Rewrite(tvm::relay::CallNode const*, tvm::RelayExpr const&)+0xf97) [0x7f4cfc977cf7] [bt] (5) /home/kk/tvm/build/libtvm.so(tvm::runtime::TypedPackedFunc<tvm::RelayExpr (tvm::relay::Call const&, tvm::runtime::Array<tvm::RelayExpr, void> const&, tvm::runtime::ObjectRef const&)>::AssignTypedLambda<tvm::RelayExpr ()(tvm::relay::Call const&, tvm::runtime::Array<tvm::RelayExpr, void> const&, tvm::runtime::ObjectRef const&)>(tvm::RelayExpr ()(tvm::relay::Call const&, tvm::runtime::Array<tvm::RelayExpr, void> const&, tvm::runtime::ObjectRef const&))::{lambda(tvm::runtime::TVMArgs const&, tvm::runtime::TVMRetValue*)#1}::operator()(tvm::runtime::TVMArgs const&, tvm::runtime::TVMRetValue*) const+0x3cf) [0x7f4cfc8ccb5f] [bt] (4) /home/kk/tvm/build/libtvm.so(tvm::RelayExpr tvm::relay::LayoutRewritertvm::relay::alter_op_layout::AlterTransformMemorizer(tvm::relay::Call const&, tvm::runtime::Array<tvm::RelayExpr, void> const&, tvm::runtime::ObjectRef const&)+0x135d) [0x7f4cfc8d4fbd] [bt] (3) /home/kk/tvm/build/libtvm.so(+0x11766b2) [0x7f4cfc8d36b2] [bt] (2) /home/kk/tvm/build/libtvm.so(tvm::runtime::TypedPackedFunc<tvm::runtime::Array<tvm::runtime::Array<tvm::tir::Layout, void>, void> (tvm::Attrs const&, tvm::runtime::Array<tvm::tir::Layout, void> const&, tvm::runtime::Array<tvm::tir::Layout, void> const&, tvm::runtime::Array<tvm::Type, void> const&)>::AssignTypedLambda<tvm::runtime::Array<tvm::runtime::Array<tvm::tir::Layout, void>, void> ()(tvm::Attrs const&, tvm::runtime::Array<tvm::tir::Layout, void> const&, tvm::runtime::Array<tvm::tir::Layout, void> const&, tvm::runtime::Array<tvm::Type, void> const&)>(tvm::runtime::Array<tvm::runtime::Array<tvm::tir::Layout, void>, void> ()(tvm::Attrs const&, tvm::runtime::Array<tvm::tir::Layout, void> const&, tvm::runtime::Array<tvm::tir::Layout, void> const&, tvm::runtime::Array<tvm::Type, void> const&))::{lambda(tvm::runtime::TVMArgs const&, tvm::runtime::TVMRetValue*)#1}::operator()(tvm::runtime::TVMArgs const&, tvm::runtime::TVMRetValue*) const+0x1ea) [0x7f4cfc5b18fa] [bt] (1) /home/kk/tvm/build/libtvm.so(tvm::relay::SliceLikeInferCorrectLayout(tvm::Attrs const&, tvm::runtime::Array<tvm::tir::Layout, void> const&, tvm::runtime::Array<tvm::tir::Layout, void> const&, tvm::runtime::Array<tvm::Type, void> const&)+0x567) [0x7f4cfc77cf17] [bt] (0) /home/kk/tvm/build/libtvm.so(+0x10063f8) [0x7f4cfc7633f8] File “/home/kk/tvm/src/relay/op/tensor/transform.cc”, line 2835 TVMError:

An internal invariant was violated during the execution of TVM. Please read TVM’s error reporting guidelines. More details can be found here: https://discuss.tvm.ai/t/error-reporting/7793.

Check failed: new_in_layouts[0]->name == new_in_layouts[1]->name (NCHW16c vs. NCHW32c) :