Rerunning this now results in the following message:
```
>>> import mxnet as mx
>>> a = mx.nd.array([1,2,3])
>>> a.astype('float16').max()
[3.]
<NDArray 1 @cpu(0)>
>>> a.astype('float16').topk()
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/home/ubuntu/topk_fp16/python/mxnet/ndarray/ndarray.py", line 189, in
__repr__
return '\n%s\n<%s %s @%s>' % (str(self.asnumpy()),
File "/home/ubuntu/topk_fp16/python/mxnet/ndarray/ndarray.py", line 1972, in
asnumpy
ctypes.c_size_t(data.size)))
File "/home/ubuntu/topk_fp16/python/mxnet/base.py", line 252, in check_call
raise MXNetError(py_str(_LIB.MXGetLastError()))
mxnet.base.MXNetError: [22:58:33]
/home/ubuntu/topk_fp16/src/operator/tensor/./ordering_op-inl.h:535: This
operation does not support float16
```
This is due to change #12250 that improves the messaging that float16 is not
supported.
We should change the tags on this issue to [Operator, Feature Request] and
remove [Bug] now that its been handled as not currently supported.
[ Full content available at:
https://github.com/apache/incubator-mxnet/issues/11156 ]
This message was relayed via gitbox.apache.org for [email protected]
