tvm-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From GitBox <...@apache.org>
Subject [GitHub] [incubator-tvm] FrozenGene commented on a change in pull request #5461: [MXNET]broadcast and logical op support
Date Thu, 30 Apr 2020 08:10:43 GMT

FrozenGene commented on a change in pull request #5461:
URL: https://github.com/apache/incubator-tvm/pull/5461#discussion_r417832135



##########
File path: python/tvm/relay/frontend/mxnet.py
##########
@@ -1712,6 +1712,33 @@ def _get_bias_requantize_scale(_inputs, _data_scale, _kernel_scale):
                 res = _op.nn.relu(res)
             return res
 
+
+def _mx_broadcast_to(inputs, attrs):
+    data = inputs[0]
+    tgt_shape = attrs.get_int_tuple("shape", [])
+
+    return _op.broadcast_to(data, tgt_shape)
+
+
+def _mx_logical_not(inputs, input_types):
+    data = inputs[0]
+    dtype = _infer_type(data).checked_type.dtype
+    data = _op.cast(data, "bool") if dtype != "bool" else data
+
+    return _op.cast(_op.logical_not(data), dtype)
+
+
+def _mx_broadcast_logical(logical_op):
+    def impl(inputs, input_types):
+        dtype0 = _infer_type(inputs[0]).checked_type.dtype

Review comment:
       code style. MXNet frontend names the variable as `dtype_0` style. Maybe we also could
consider more meaningful name, like `lhs_dtype`.




----------------------------------------------------------------
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



Mime
View raw message