!3013 fix tensor bool and float scalar

Merge pull request !3013 from amongo/fix_tensor_bool_add_float

mindspore/ccsrc/operator/composite/do_signature.cc

@@ -106,6 +106,8 @@ TypeId GetMaxTypeId(const abstract::AbstractBasePtrList &args_spec_list, std::ve
   TypeId max_type_id = kTypeUnknown;
   size_t max_type_number = 0;
   bool has_int8 = false;
+  bool has_scalar_int32 = false;
+  bool has_scalar_float32 = false;
   for (const auto &index : indices) {
     TypeId arg_type_id = kTypeUnknown;
     TypeId arg_type = kTypeUnknown;
@@ -114,6 +116,11 @@ TypeId GetMaxTypeId(const abstract::AbstractBasePtrList &args_spec_list, std::ve
       continue;
     }
     if (arg_type != kObjectTypeTensorType) {
+      if (arg_type_id == kNumberTypeInt32) {
+        has_scalar_int32 = true;
+      } else if (arg_type_id == kNumberTypeFloat32) {
+        has_scalar_float32 = true;
+      }
       continue;
     }
     auto it = type_map.find(arg_type_id);
@@ -135,6 +142,17 @@ TypeId GetMaxTypeId(const abstract::AbstractBasePtrList &args_spec_list, std::ve
   if (max_type_id == kNumberTypeUInt8 && has_int8 == true) {
     max_type_id = kNumberTypeInt16;
   }
+  // if bool is the max type, see if there is scalar input
+  // if so, it means that max is bool tensor, use scalar type instead.
+  // for example: Tensor([True, True]) * 2, expect result is Tensor([2, 2])
+  if (max_type_id == kNumberTypeBool) {
+    if (has_scalar_int32) {
+      max_type_id = kNumberTypeInt32;
+    }
+    if (has_scalar_float32) {
+      max_type_id = kNumberTypeFloat32;
+    }
+  }
   return max_type_id;
 }
 

tests/st/ops/ascend/test_autocast.py

@@ -246,3 +246,21 @@ def test_tensor_auto_cast():
         bnet(t_fp32)
     with pytest.raises(TypeError):
         bnet(t_fp64)
+def test_bool_tensor_and_float():
+    context.set_context(mode=context.GRAPH_MODE)
+    t_bool = Tensor(np.ones([2, 1, 2, 2]).astype(np.bool), mstype.bool_)
+    t_int32 = Tensor(np.ones([2, 1, 2, 2]), mstype.int32)
+    t_fp16 = Tensor(np.ones([2, 1, 2, 2]), mstype.float16)
+    t_fp32 = Tensor(np.ones([2, 1, 2, 2]), mstype.float32)
+    net = TensorFPAutoCast()
+    out = net(t_bool)
+    assert out.dtype == mstype.float32
+    net = TensorIntAutoCast()
+    out = net(t_bool)
+    assert out.dtype == mstype.int32
+    out = net(t_fp16)
+    assert out.dtype == mstype.float16
+    out = net(t_fp32)
+    assert out.dtype == mstype.float32
+    out = net(t_int32)
+    assert out.dtype == mstype.int32