You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@mxnet.apache.org by GitBox <gi...@apache.org> on 2018/10/12 21:17:15 UTC

[GitHub] anirudhacharya commented on a change in pull request #12812: [MXNET-886] ONNX export: HardSigmoid, Less, Greater, Equal, InstanceNorm

anirudhacharya commented on a change in pull request #12812: [MXNET-886] ONNX export: HardSigmoid, Less, Greater, Equal, InstanceNorm
URL: https://github.com/apache/incubator-mxnet/pull/12812#discussion_r224913029
 
 

 ##########
 File path: tests/python-pytest/onnx/export/mxnet_export_test.py
 ##########
 @@ -179,8 +180,80 @@ def test_model_accuracy(model_name, input_shape):
         npt.assert_equal(expected.shape, actual.shape)
         npt.assert_almost_equal(expected, actual, decimal=3)
 
+def test_greater():
+    """Test for logical greater in onnx operators."""
+    input1 = np.array([[ 1,  1,  1], [ 1,  1,  1]]).astype("float32")
+    input2 = np.array([[ 0], [ 1]]).astype("float32")
+    op = np.array([[ 1,  1,  1], [ 0,  0,  0]]).astype("float32")
+    inputs = [helper.make_tensor_value_info("input1", TensorProto.FLOAT, shape=np.shape(input1)),
+              helper.make_tensor_value_info("input2", TensorProto.FLOAT, shape=np.shape(input2))]
+
+    outputs = [helper.make_tensor_value_info("output", TensorProto.FLOAT, shape=np.shape(op))]
+
+    nodes = [helper.make_node("Greater", ["input1", "input2"], ["output"])]
+
+    graph = helper.make_graph(nodes,
+                              "greater_test",
+                              inputs,
+                              outputs)
+
+    greater_model = helper.make_model(graph)
+
+    bkd_rep = backend.prepare(greater_model)
+    output = bkd_rep.run([input1, input2])
+    npt.assert_almost_equal(output[0], op)
+
+def test_equal():
+    """Test for equal in onnx operators."""
+    input1 = np.array([[ 1,  1,  1], [ 1,  1,  1]]).astype("float32")
+    input2 = np.array([[ 0], [ 1]]).astype("float32")
+    op = np.array([[ 0,  0,  0], [ 1,  1,  1]]).astype("float32")
+    inputs = [helper.make_tensor_value_info("input1", TensorProto.FLOAT, shape=np.shape(input1)),
+              helper.make_tensor_value_info("input2", TensorProto.FLOAT, shape=np.shape(input2))]
+
+    outputs = [helper.make_tensor_value_info("output", TensorProto.FLOAT, shape=np.shape(op))]
+
+    nodes = [helper.make_node("Equal", ["input1", "input2"], ["output"])]
+
+    graph = helper.make_graph(nodes,
+                              "equal_test",
+                              inputs,
+                              outputs)
+
+    equal_model = helper.make_model(graph)
+
+    bkd_rep = backend.prepare(equal_model)
+    output = bkd_rep.run([input1, input2])
+    npt.assert_almost_equal(output[0], op)
+
+def test_lesser():
+    """Test for lesser in onnx operators."""
+    input1 = np.array([[ 1,  1,  1], [ 1,  1,  1]]).astype("float32")
+    input2 = np.array([[ 0], [ 1]]).astype("float32")
+    op = np.array([[ 0,  0,  0], [ 0,  0,  0]]).astype("float32")
+    inputs = [helper.make_tensor_value_info("input1", TensorProto.FLOAT, shape=np.shape(input1)),
+              helper.make_tensor_value_info("input2", TensorProto.FLOAT, shape=np.shape(input2))]
+
+    outputs = [helper.make_tensor_value_info("output", TensorProto.FLOAT, shape=np.shape(op))]
+
+    nodes = [helper.make_node("Less", ["input1", "input2"], ["output"])]
+
+    graph = helper.make_graph(nodes,
+                              "less_test",
+                              inputs,
+                              outputs)
+
+    lesser_model = helper.make_model(graph)
+
+    bkd_rep = backend.prepare(lesser_model)
+    output = bkd_rep.run([input1, input2])
+    npt.assert_almost_equal(output[0], op)
 
 if __name__ == '__main__':
+    test_greater()
 
 Review comment:
   do not call each of the test methods separately. These tests are written based on pytest framework. You can say `unittest.main()`. pytest at runtime will collect the tests that need to be run.

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on 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


With regards,
Apache Git Services