You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@tvm.apache.org by GitBox <gi...@apache.org> on 2020/09/25 05:29:31 UTC
[GitHub] [incubator-tvm] rubick1896 opened a new issue #6556: [New Op request] _sample_normal and _sample_gamma from mxnet 1.7.0
rubick1896 opened a new issue #6556:
URL: https://github.com/apache/incubator-tvm/issues/6556
I got the following message when trying to compile DeepAR model from Gluonts
`
---------------------------------------------------------------------------
OpNotImplemented Traceback (most recent call last)
<ipython-input-3-9838f83506d7> in <module>
----> 1 mod, params = relay.frontend.from_mxnet(predictor.prediction_net, shape_dict)
~/.local/lib/python3.7/site-packages/tvm-0.7.dev1-py3.7-linux-x86_64.egg/tvm/relay/frontend/mxnet.py in from_mxnet(symbol, shape, dtype, arg_params, aux_params)
2596 sym = mx.sym.Group(sym)
2597 shape, dtype = _update_shape_dtype(shape, dtype, params)
-> 2598 func = _from_mxnet_impl(sym, shape, dtype, params, mod)
2599 elif isinstance(symbol, mx.gluon.Block):
2600 raise NotImplementedError("Only Hybrid Blocks are supported now.")
~/.local/lib/python3.7/site-packages/tvm-0.7.dev1-py3.7-linux-x86_64.egg/tvm/relay/frontend/mxnet.py in _from_mxnet_impl(symbol, shape_dict, dtype_info, params, mod)
2472 msg = '\n'.join(['{}: {}'.format(op_name, cnt) for op_name, cnt in unsupported.items()])
2473 raise tvm.error.OpNotImplemented(
-> 2474 'One or more operators are not supported in frontend MXNet:\n{}'.format(msg))
2475
2476 for nid, node in enumerate(jnodes):
OpNotImplemented: One or more operators are not supported in frontend MXNet:
_sample_gamma: 10
_sample_normal: 10
`
DeepAR: https://gluon-ts.mxnet.io/api/gluonts/gluonts.model.deepar.html
TVM version: 0.7.dev1
Is it possible to support these operations in the near future?
----------------------------------------------------------------
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:
users@infra.apache.org