Hi @tqchen
I was just trying to import a simple “Upsample” layer via the TVM frontend and it throws the following assertion. I am using ONNX 1.4.1 and TVM v.5. Can you please help me…
import nnvm
import tvm
import onnx
import numpy as np
onnx_graph = onnx.load(‘export_upsample.onnx’)
sym, params = nnvm.frontend.from_onnx(onnx_graph)
File “/Users/pallabdatta/nnvm_tvm_npcompiler/tvm/nnvm/python/nnvm/frontend/onnx.py”, line 975, in from_onnx
sym, params = g.from_onnx(graph, opset)
File “/Users/pallabdatta/nnvm_tvm_npcompiler/tvm/nnvm/python/nnvm/frontend/onnx.py”, line 830, in from_onnx
op = self._convert_operator(op_name, inputs, attr, opset)
File “/Users/pallabdatta/nnvm_tvm_npcompiler/tvm/nnvm/python/nnvm/frontend/onnx.py”, line 931, in _convert_operator
sym = convert_map[op_name](inputs, attrs, self._params)
File “/Users/pallabdatta/nnvm_tvm_npcompiler/tvm/nnvm/python/nnvm/frontend/onnx.py”, line 393, in _impl_v7
assert len(scales) == 4 and scales[0] == 1.0 and scales[1] == 1.0 and scales[2] == scales[3]
TypeError: object of type ‘NoneType’ has no len()
>>>