I’m using relay.frontend.from_pytorch
to convert pytorch model, but raise error like this:
Traceback (most recent call last):
File "tune_network_cuda_test.py", line 200, in <module>
mod, params, input_shape, output_shape = get_local_network("pt", network, batch_size, shape, layout)
File "tune_network_cuda_test.py", line 97, in get_local_network
mod, params = relay.frontend.from_pytorch(torch_model, shape_info)
File "/wen/yincw/open_source/tvm/python/tvm/relay/frontend/pytorch.py", line 3285, in from_pytorch
converter.report_missing_conversion(op_names)
File "/wen/yincw/open_source/tvm/python/tvm/relay/frontend/pytorch.py", line 2545, in report_missing_conversion
raise NotImplementedError(msg)
NotImplementedError: The following operators are not implemented: ['aten::im2col']
and another similar error:
Traceback (most recent call last):
File "tune_network_cuda_test.py", line 209, in <module>
mod, params, input_shape, output_shape = get_local_network("pt", network, batch_size, shape, layout)
File "tune_network_cuda_test.py", line 106, in get_local_network
mod, params = relay.frontend.from_pytorch(torch_model, shape_info)
File "/wen/yincw/open_source/tvm/python/tvm/relay/frontend/pytorch.py", line 3285, in from_pytorch
converter.report_missing_conversion(op_names)
File "/wen/yincw/open_source/tvm/python/tvm/relay/frontend/pytorch.py", line 2545, in report_missing_conversion
raise NotImplementedError(msg)
NotImplementedError: The following operators are not implemented: ['aten::grid_sampler']
pytorch=1.3 tvm=0.8 dev0
Could anyone help?