652994331 opened a new issue #6139: URL: https://github.com/apache/incubator-tvm/issues/6139
Hi, guys, i built the tvm 0.6.1 and build the python bind successfully, i was folllowing this (https://gist.github.com/icemelon9/860d3d2c9566d6f69fa8112840dd95c1) to do a bert inference. However, i got this problem: (tvm) [qizhouhuang@GPU-10-252-192-90 apache-tvm-src-v0.6.1.rc1-incubating]$ python3 test_bert.py MXNet latency for batch 1 and seq length 128: 114.93 ms Traceback (most recent call last): File "test_bert.py", line 86, in <module> mod, params = relay.frontend.from_mxnet(model, shape_dict) File "/opt/cephfs1/asr/users/qizhou.huang/.local/lib/python3.6/site-packages/tvm-0.6.0-py3.6-linux-x86_64.egg/tvm/relay/frontend/mxnet.py", line 1427, in from_mxnet func = _from_mxnet_impl(sym, shape, dtype, mod) File "/opt/cephfs1/asr/users/qizhou.huang/.local/lib/python3.6/site-packages/tvm-0.6.0-py3.6-linux-x86_64.egg/tvm/relay/frontend/mxnet.py", line 1340, in _from_mxnet_impl 'Operator {} is not supported in frontend MXNet.'.format(op_name)) tvm.error.OpNotImplemented: Operator _contrib_arange_like is not supported in frontend MXNet. I installed my mxnet using pip install mxnet, it seems the one got installed's Collecting mxnet Using cached mxnet-1.6.0-py2.py3-none-any.whl (68.7 MB) could u please help me out,thank you. ---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected]
