[VTA][NNVM] Error while inferencing on PYNQ-Z2 target

Questions
1

Hello everyone.
I’m attempting to run tiny yolo v2 on PYNQ-Z2 target.
Since I have encountered errors when running relay models on VTA(simulated environment also),
I tried to run NNVM models first.
target sim runs successfully with the same code.

So my question is what caused these errors?
I would happy if you have any suggestions.
Thanks,

create graph
graph created
Start inference
Traceback (most recent call last):
  File "tinyyolo.py", line 379, in <module>
    m.run()
  File "/tvm/python/tvm/contrib/graph_runtime.py", line 151, in run
    self._run()
  File "/tvm/python/tvm/_ffi/_ctypes/function.py", line 185, in __call__
    ctypes.byref(ret_val), ctypes.byref(ret_tcode)))
  File "/tvm/python/tvm/_ffi/base.py", line 71, in check_call
    raise TVMError(py_str(_LIB.TVMGetLastError()))
tvm._ffi.base.TVMError: Except caught from RPC call: [02:33:11] /tvm/vta/src/runtime.cc:69: Check failed: data != nullptr

Stack trace returned 10 entries:
[bt] (0) /tvm/vta/python/vta/../../../build/libvta.so(+0x8b96) [0xb1527b96]
[bt] (1) /tvm/vta/python/vta/../../../build/libvta.so(VTABufferAlloc+0x195) [0xb1522a12]
[bt] (2) /tvm/build/libtvm.so(tvm::runtime::WorkspacePool::AllocWorkspace(DLContext, unsigned int)+0x119) [0xb58a5a3e]
[bt] (3) /tvm/vta/python/vta/../../../build/libvta.so(+0x2a4c) [0xb1521a4c]
[bt] (4) /tmp/tmp5x7iakmg/graphlib.o.so(+0x1b210) [0xb0b5c210]
[bt] (5) /tmp/tmp5x7iakmg/graphlib.o.so(fuse_conv2d_7+0x41c) [0xb0b5bda0]
[bt] (6) /home/makky/tvm/build/libtvm.so(+0x52add6) [0xb5895dd6]
[bt] (7) /home/makky/tvm/build/libtvm.so(+0x5538d6) [0xb58be8d6]
[bt] (8) /home/makky/tvm/build/libtvm.so(+0x552708) [0xb58bd708]
[bt] (9) /home/makky/tvm/build/libtvm.so(+0x547ec6) [0xb58b2ec6]

This error also appears depending on the timing.

create graph
Traceback (most recent call last):
  File "tinyyolo.py", line 364, in <module>
    m = graph_runtime.create(graph, lib, ctx)
  File "/tvm/python/tvm/contrib/graph_runtime.py", line 40, in create
    return GraphModule(fcreate(graph_json_str, hmod, *device_type_id))
  File "/tvm/python/tvm/_ffi/_ctypes/function.py", line 185, in __call__
    ctypes.byref(ret_val), ctypes.byref(ret_tcode)))
  File "/tvm/python/tvm/_ffi/base.py", line 71, in check_call
    raise TVMError(py_str(_LIB.TVMGetLastError()))
tvm._ffi.base.TVMError: Except caught from RPC call: [02:56:38] /tvm/vta/src/runtime.cc:69: Check failed: data != nullptr

Stack trace returned 10 entries:
[bt] (0) /tvm/vta/python/vta/../../../build/libvta.so(+0x8b96) [0xb15aab96]
[bt] (1) /tvm/vta/python/vta/../../../build/libvta.so(VTABufferAlloc+0x195) [0xb15a5a12]
[bt] (2) /tvm/build/libtvm.so(tvm::runtime::NDArray::Empty(std::vector<long long, std::allocator<long long> >, DLDataType, DLContext)+0x131) [0xb591d146]
[bt] (3) /tvm/build/libtvm.so(+0x5546e6) [0xb59426e6]
[bt] (4) /tvm/build/libtvm.so(+0x554c96) [0xb5942c96]
[bt] (5) /tvm/build/libtvm.so(+0x554e14) [0xb5942e14]
[bt] (6) /tvm/build/libtvm.so(+0x5552e6) [0xb59432e6]
[bt] (7) /tvm/build/libtvm.so(+0x547ec6) [0xb5935ec6]
[bt] (8) /tvm/build/libtvm.so(+0x54bba4) [0xb5939ba4]
[bt] (9) /tvm/build/libtvm.so(+0x54c848) [0xb593a848]

Here is the source code and model data attached.