ZheyuYe opened a new issue #18766: URL: https://github.com/apache/incubator-mxnet/issues/18766
## Description I tried to run Horovod on the latest Mxnet master branch and it failed as below. ### Steps to reproduce Horovod was installed by `HOROVOD_GPU_OPERATIONS=NCCL pip3 install --no-cache-dir horovod ` [A gist exmaple](https://gist.github.com/ZheyuYe/f59b33c20a0b6fdcbf471af3e2d5ac64) revised by [offical mnist example](https://github.com/apache/incubator-mxnet/blob/master/example/distributed_training-horovod/gluon_mnist.py) was executed with `mpirun -np 4 -H localhost:4 -bind-to none -map-by slot python3 mxnet_mnist.py` Please point out the steps I missed if I did't run this example properly. ## Outputs ```bash File "/home/ubuntu/.local/lib/python3.6/site-packages/horovod/mxnet/__init__.py", line 154, in broadcast_parameters broadcast_(tensor, root_rank, name=str(name)) File "/home/ubuntu/.local/lib/python3.6/site-packages/horovod/mxnet/mpi_ops.py", line 232, in broadcast_ c_in = tensor.handle AttributeError: 'Parameter' object has no attribute 'handle' Traceback (most recent call last): File "gluon_mnist.py", line 144, in <module> hvd.broadcast_parameters(params, root_rank=0) File "/home/ubuntu/.local/lib/python3.6/site-packages/horovod/mxnet/__init__.py", line 154, in broadcast_parameters broadcast_(tensor, root_rank, name=str(name)) File "/home/ubuntu/.local/lib/python3.6/site-packages/horovod/mxnet/mpi_ops.py", line 232, in broadcast_ c_in = tensor.handle AttributeError: 'Parameter' object has no attribute 'handle' -------------------------------------------------------------------------- Primary job terminated normally, but 1 process returned a non-zero exit code. Per user-direction, the job has been aborted. -------------------------------------------------------------------------- -------------------------------------------------------------------------- mpirun detected that one or more processes exited with non-zero status, thus causing the job to be terminated. The first process to do so was: Process name: [[28180,1],1] Exit code: 1 -------------------------------------------------------------------------- ``` ## Comments @leezu @sxjscience ---------------------------------------------------------------- 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: us...@infra.apache.org
