VTA shows errors while executing the YOLOv3-tiny example

Troubleshooting
#1

Hello, I’m trying to run the YOLOv3-tiny example (tvm/vta/tutorials/frontend/legacy/deploy_detection.py). The script terminates with the error message below.

System

Running under Ubuntu 18.04.5 LTS and Docker images created with Dockerfile.ci_cpu. Version: 0.8.dev0

Error message

## tvm._ffi.base.TVMError: Traceback (most recent call last):
[bt] (7) /workspace/.local/lib/python3.6/site-packages/tvm-0.8.dev238+gdc7220447-py3.6-linux-x86_64.egg/tvm/libtvm.so(TVMFuncCall+0x65) [0x7f1f51478b75]
[bt] (6) /workspace/.local/lib/python3.6/site-packages/tvm-0.8.dev238+gdc7220447-py3.6-linux-x86_64.egg/tvm/libtvm.so(+0x6496c2) [0x7f1f508f56c2]
[bt] (5) /workspace/.local/lib/python3.6/site-packages/tvm-0.8.dev238+gdc7220447-py3.6-linux-x86_64.egg/tvm/libtvm.so(tvm::transform::ModulePassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const+0x1b7) [0x7f1f508f4f77]
[bt] (4) /workspace/.local/lib/python3.6/site-packages/tvm-0.8.dev238+gdc7220447-py3.6-linux-x86_64.egg/tvm/libtvm.so(+0xfd19cf) [0x7f1f5127d9cf]
[bt] (3) /workspace/.local/lib/python3.6/site-packages/tvm-0.8.dev238+gdc7220447-py3.6-linux-x86_64.egg/tvm/libtvm.so(+0xfd0c25) [0x7f1f5127cc25]
[bt] (2) /workspace/.local/lib/python3.6/site-packages/tvm-0.8.dev238+gdc7220447-py3.6-linux-x86_64.egg/tvm/libtvm.so(tvm::relay::TypeInferencer::Infer(tvm::GlobalVar, tvm::relay::Function)+0x67) [0x7f1f5127c1d7]
[bt] (1) /workspace/.local/lib/python3.6/site-packages/tvm-0.8.dev238+gdc7220447-py3.6-linux-x86_64.egg/tvm/libtvm.so(tvm::relay::TypeSolver::Solve()+0xd39) [0x7f1f51120599]
[bt] (0) /workspace/.local/lib/python3.6/site-packages/tvm-0.8.dev238+gdc7220447-py3.6-linux-x86_64.egg/tvm/libtvm.so(+0xe70732) [0x7f1f5111c732]
[bt] (8) /workspace/.local/lib/python3.6/site-packages/tvm-0.8.dev238+gdc7220447-py3.6-linux-x86_64.egg/tvm/libtvm.so(+0x6496c2) [0x7f1f508f56c2]
[bt] (7) /workspace/.local/lib/python3.6/site-packages/tvm-0.8.dev238+gdc7220447-py3.6-linux-x86_64.egg/tvm/libtvm.so(tvm::transform::ModulePassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const+0x1b7) [0x7f1f508f4f77]
[bt] (6) /workspace/.local/lib/python3.6/site-packages/tvm-0.8.dev238+gdc7220447-py3.6-linux-x86_64.egg/tvm/libtvm.so(+0xfd19cf) [0x7f1f5127d9cf]
[bt] (5) /workspace/.local/lib/python3.6/site-packages/tvm-0.8.dev238+gdc7220447-py3.6-linux-x86_64.egg/tvm/libtvm.so(+0xfd0c25) [0x7f1f5127cc25]
[bt] (4) /workspace/.local/lib/python3.6/site-packages/tvm-0.8.dev238+gdc7220447-py3.6-linux-x86_64.egg/tvm/libtvm.so(tvm::relay::TypeInferencer::Infer(tvm::GlobalVar, tvm::relay::Function)+0x67) [0x7f1f5127c1d7]
[bt] (3) /workspace/.local/lib/python3.6/site-packages/tvm-0.8.dev238+gdc7220447-py3.6-linux-x86_64.egg/tvm/libtvm.so(tvm::relay::TypeSolver::Solve()+0x36d) [0x7f1f5111fbcd]
[bt] (2) /workspace/.local/lib/python3.6/site-packages/tvm-0.8.dev238+gdc7220447-py3.6-linux-x86_64.egg/tvm/libtvm.so(std::_Function_handler<void (tvm::runtime::TVMArgs,     tvm::runtime::TVMRetValue*), tvm::runtime::TypedPackedFunc<bool (tvm::runtime::Array<tvm::Type, void> const&, int, tvm::Attrs const&, tvm::TypeReporter const&)>::AssignTypedLambda<bool ( *)(tvm::runtime::Array<tvm::Type, void> const&, int, tvm::Attrs const&, tvm::TypeReporter const&)>(bool (* )(tvm::runtime::Array<tvm::Type, void> const&, int, tvm::Attrs const&, tvm::TypeReporter const&))::{lambda(tvm::runtime::TVMArgs const&, tvm::runtime::TVMRetValue*)[#1](https://github.com/apache/tvm/issues/1)}>::_M_invoke(std::_Any_data const&, tvm::runtime::TVMArgs&&, tvm::runtime::TVMRetValue*&&)+0x7d7) [0x7f1f50d8d0e7]
    [bt] (1) /workspace/.local/lib/python3.6/site-packages/tvm-0.8.dev238+gdc7220447-py3.6-linux-x86_64.egg/tvm/libtvm.so(tvm::relay::ReshapeRel(tvm::runtime::Array<tvm::Type, void> const&, int, tvm::Attrs const&, tvm::TypeReporter const&)+0x709) [0x7f1f5104f759]
    [bt] (0) /workspace/.local/lib/python3.6/site-packages/tvm-0.8.dev238+gdc7220447-py3.6-linux-x86_64.egg/tvm/libtvm.so(+0xd75d12) [0x7f1f51021d12]
    File "/workspace/tvm/src/relay/analysis/type_solver.cc", line 621
    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: false == false: [16:00:03] /workspace/tvm/src/relay/op/tensor/transform.cc:632:

## 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: oshape_sum == data_shape_sum (172380 vs. 173056) : Input tensor shape and 
reshaped shape are not compatible

TVM/VTA is new to me, help is really appreciated. Thanks!

#2

Hi! I managed to solve this by removing two lines from the graph_pack step:

            mod = graph_pack(
                mod["main"],
                env.BATCH,
                env.BLOCK_OUT,
                env.WGT_WIDTH,
                start_name=pack_dict[MODEL_NAME][0],
                stop_name=pack_dict[MODEL_NAME][1]
      #         start_name_idx=pack_dict[MODEL_NAME][2], 
      #         stop_name_idx=pack_dict[MODEL_NAME][3],
            )

Hope this helps