Issue: [ONNX][OP] IndexError: list index out of range


#1

Hi,

I am facing error while compiling ONNX model with GEMM operation as below.

Could anyone help me with the issue.

test script:

import nnvm
import tvm
import onnx
import numpy as np
import tensorflow as tf
onnx_model = onnx.load(‘gemm3.onnx’)
sym, params = nnvm.frontend.from_onnx(onnx_model)
print(“debug symbols:”,sym.debug_str())
print("\n loading in nnvm done ")

Error Log :
Traceback (most recent call last):
File “tvm_opencl_opt0gemm.py”, line 13, in
sym, params = nnvm.frontend.from_onnx(onnx_model)
File “/home/ubuntu/tvm_opencl/tvm/nnvm/python/nnvm/frontend/onnx.py”, line 967, in from_onnx
sym, params = g.from_onnx(graph, opset)
File “/home/ubuntu/tvm_opencl/tvm/nnvm/python/nnvm/frontend/onnx.py”, line 822, in from_onnx
op = self._convert_operator(op_name, inputs, attr, opset)
File “/home/ubuntu/tvm_opencl/tvm/nnvm/python/nnvm/frontend/onnx.py”, line 923, in _convert_operator
sym = convert_map[op_name](inputs, attrs, self._params)
File “/home/ubuntu/tvm_opencl/tvm/nnvm/python/nnvm/frontend/onnx.py”, line 183, in _impl_v1
channels = infer_channels(inputs[1], params, not transB)
File “/home/ubuntu/tvm_opencl/tvm/nnvm/python/nnvm/frontend/onnx_caffe2_utils.py”, line 33, in infer_channels
channels = out_shapes[0][0] if not transpose else out_shapes[0][1]
IndexError: list index out of range


#2

@tqchen could you please help here


#3

@alexeyr @srkreddy1238 could you please help with the issue
PFA graphgemm%20graph