modifiy sign_op4 and test_sign

Author: mindest

tests/test_backend.py

@@ -1068,11 +1068,15 @@ def test_sign_int(self):
         self._run_test_case([_OUTPUT], {_INPUT: x_val})
 
     def test_sign(self):
-        x_val = np.array([1.0, 2.0, 0.0, -1.0, 0.0, -2.0], dtype=np.float32).reshape((2, 3))
-        x = tf.placeholder(tf.float32, [2, 3], name=_TFINPUT)
-        x_ = tf.sign(x)
-        _ = tf.identity(x_, name=_TFOUTPUT)
-        self._run_test_case([_OUTPUT], {_INPUT: x_val})
+        x_val1 = np.array([1, 2, 0, -1, 0, -2], dtype=np.int32).reshape((2, 3))
+        x_val2 = np.array([1, 2, 0, -1, 0, -2], dtype=np.int64).reshape((2, 3))
+        x_val3 = np.array([1.0, 2.0, 0.0, -1.0, 0.0, -2.0], dtype=np.float32).reshape((2, 3))
+        for x_val in [x_val1, x_val2, x_val3]:
+            x = tf.placeholder(x_val.dtype, x_val.shape, name=_TFINPUT)
+            x_ = tf.sign(x)
+            _ = tf.identity(x_, name=_TFOUTPUT)
+            self._run_test_case([_OUTPUT], {_INPUT: x_val})
+            tf.reset_default_graph()
 
     def test_onehot0(self):
         x_val = np.array([0, 1, 2], dtype=np.int32)

tf2onnx/tfonnx.py

@@ -648,12 +648,16 @@ def sign_op4(ctx, node, name, args):
     node_dtype = ctx.get_dtype(node.output[0])
     utils.make_sure(node_dtype, "Dtype of {} is None".format(node.name))
     if node_dtype in [onnx_pb.TensorProto.COMPLEX64, onnx_pb.TensorProto.COMPLEX128]:
-        raise ValueError("dtype " + node_dtype + " is not supported in onnx for now")
-    input_tensor_type = utils.map_onnx_to_numpy_type(node_dtype)
+        raise ValueError("dtype " + str(node_dtype) + " is not supported in onnx for now")
     zero_name = utils.make_name("{}_zero".format(node.name))
-    ctx.make_const(zero_name, np.array(0, dtype=input_tensor_type))
-    greater_node = ctx.make_node("Greater", [node.input[0], zero_name])
-    less_node = ctx.make_node("Less", [node.input[0], zero_name])
+    ctx.make_const(zero_name, np.array(0, dtype=utils.ONNX_TO_NUMPY_DTYPE[1]))
+    if node_dtype in [onnx_pb.TensorProto.INT32, onnx_pb.TensorProto.INT64]:
+        cast_node_0 = ctx.make_node("Cast", [node.input[0]], {"to": 1})
+        greater_node = ctx.make_node("Greater", [cast_node_0.output[0], zero_name])
+        less_node = ctx.make_node("Less", [cast_node_0.output[0], zero_name])
+    else:
+        greater_node = ctx.make_node("Greater", [node.input[0], zero_name])
+        less_node = ctx.make_node("Less", [node.input[0], zero_name])
     cast_node_1 = ctx.make_node("Cast", [greater_node.output[0]], {"to": node_dtype})
     cast_node_2 = ctx.make_node("Cast", [less_node.output[0]], {"to": node_dtype})