Enhance upsample operator to adapt onnx opset version 9 for nnvm comp… (apache#2968)

* Enhance upsample operator to adapt onnx opset version 9 for nnvm compiler

* Add upsample test case for newer opset in nnvm

* re-trigger the CI

Author: gemfield

nnvm/python/nnvm/frontend/onnx.py

@@ -404,8 +404,14 @@ class Upsample(OnnxOpConverter):
     """
 
     @classmethod
-    def _impl_v7(cls, inputs, attr, params):
+    def _impl_v9(cls, inputs, attr, params):
         scales = attr.get('scales')
+        if not scales:
+            #Here we are going to higher OPSET version.
+            assert len(inputs) == 2, "Upsample op take 2 inputs, {} given".format(len(inputs))
+            input_name = inputs[1].list_input_names()[0]
+            scales = params[input_name].asnumpy()
+            inputs = inputs[:1]
         assert len(scales) == 4 and scales[0] == 1.0 and scales[1] == 1.0 and scales[2] == scales[3]
         mode = attr.get('mode')
         if mode == b'nearest':

nnvm/tests/python/frontend/onnx/test_forward.py

@@ -438,9 +438,41 @@ def _test_upsample_bilinear():
         tvm_out = get_tvm_output(model, in_array, target, ctx, out_shape, 'float32')
         tvm.testing.assert_allclose(out_array, tvm_out, rtol=1e-5, atol=1e-5)
 
+def _test_upsample_bilinear_opset9():
+    scale = 2
+    in_shape = (1, 1, 3, 3)
+    out_shape = (1, 1, 3*scale, 3*scale)
+    y = helper.make_node("Upsample", ['in','scales'], ['out'], mode='linear')
+    scales=[1.0, 1.0, 2.0, 2.0]
+    in_array = np.random.uniform(size=in_shape).astype(np.float32)
+    out_array = topi.testing.bilinear_resize_python(in_array, (3*scale, 3*scale), "NCHW")
+
+    ref_array = np.array(scales)
+    ref_node = helper.make_node('Constant',
+                                 inputs=[],
+                                 outputs=['scales'],
+                                 value=onnx.helper.make_tensor(name = 'const_tensor',
+                                                               data_type = TensorProto.FLOAT,
+                                                               dims = ref_array.shape,
+                                                               vals = ref_array.flatten().astype(float)))
+
+    graph = helper.make_graph([ref_node, y],
+                              'upsample_bilinear_opset9_test',
+                              inputs = [helper.make_tensor_value_info("in", TensorProto.FLOAT, list(in_shape))],
+                              outputs = [helper.make_tensor_value_info("out", TensorProto.FLOAT, list(out_shape))])
+
+    model = helper.make_model(graph, producer_name='upsample_bilinear_opset9_test')
+    inputs = []
+    inputs.append(in_array)
+
+    for target, ctx in ctx_list():
+        tvm_out = get_tvm_output(model, inputs, target, ctx, out_shape, 'float32')
+        tvm.testing.assert_allclose(out_array, tvm_out, rtol=1e-5, atol=1e-5)
+
 def test_upsample():
     _test_upsample_nearest()
     _test_upsample_bilinear()
+    _test_upsample_bilinear_opset9()
 
 def _test_softmax(inshape, axis):
     opname = 'Softmax'