!11396 fix Matmul ops dynamic shape problems

From: @TFbunny
Reviewed-by: @robingrosman
Signed-off-by:

mindspore/core/abstract/prim_maths.cc

@@ -339,32 +339,50 @@ AbstractBasePtr InferImplMatMul(const AnalysisEnginePtr &, const PrimitivePtr &p
   auto y = CheckArg<AbstractTensor>(op_name, args_spec_list, 1);
   MS_EXCEPTION_IF_NULL(y);
   MS_EXCEPTION_IF_NULL(y->shape());
-  if (x->shape()->shape().size() != 2 || y->shape()->shape().size() != 2) {
+  auto x_shp = x->shape()->shape();
+  auto y_shp = y->shape()->shape();
+  if (x_shp.size() != 2 || y_shp.size() != 2) {
     MS_LOG(EXCEPTION) << "MatMul inputs should have the same dimension size and equal to 2.";
   }
-  ValuePtr TAptr = primitive->GetAttr("transpose_a");
-  ValuePtr TBptr = primitive->GetAttr("transpose_b");
-  bool TA = GetValue<bool>(TAptr);
-  bool TB = GetValue<bool>(TBptr);
+  ValuePtr transpose_a_ptr = primitive->GetAttr("transpose_a");
+  ValuePtr transpose_b_ptr = primitive->GetAttr("transpose_b");
+  bool transpose_a = GetValue<bool>(transpose_a_ptr);
+  bool transpose_b = GetValue<bool>(transpose_b_ptr);
   ShapeVector x_min_shape = x->shape()->min_shape();
   ShapeVector x_max_shape = x->shape()->max_shape();
   ShapeVector y_min_shape = y->shape()->min_shape();
   ShapeVector y_max_shape = y->shape()->max_shape();
-  (void)CheckMinMaxShape(x->shape()->shape(), &x_min_shape, &x_max_shape);
-  (void)CheckMinMaxShape(y->shape()->shape(), &y_min_shape, &y_max_shape);
+  (void)CheckMinMaxShape(x_shp, &x_min_shape, &x_max_shape);
+  (void)CheckMinMaxShape(y_shp, &y_min_shape, &y_max_shape);
+  // Additional check for dynamic shape
+  // Last infer will be real shape values
+  bool x_not_dyn = std::all_of(x_shp.begin(), x_shp.end(), [](int64_t value) { return value != Shape::SHP_ANY; });
+  bool y_not_dyn = std::all_of(y_shp.begin(), y_shp.end(), [](int64_t value) { return value != Shape::SHP_ANY; });
+  if (x_not_dyn && y_not_dyn) {
+    auto x_col = x_shp[(transpose_a ? 0 : 1)];
+    auto y_row = y_shp[(transpose_b ? 1 : 0)];
+    if (x_col != y_row) {
+      MS_LOG(EXCEPTION) << "MatMul shape error, got x_col: " << x_col << ", y_row: " << y_row
+                        << ". In MatMul x_col and y_row should be equal.";
+    }
+  }
   ShapeVector ret_shape;
   ShapeVector ret_min_shape;
   ShapeVector ret_max_shape;
-  auto make_shape = [&TA, &TB](ShapeVector &output, const ShapeVector xshp, const ShapeVector yshp) -> void {
-    output.push_back(xshp[(TA ? 1 : 0)]);
-    output.push_back(yshp[(TB ? 0 : 1)]);
+  auto make_shape = [&transpose_a, &transpose_b](ShapeVector &output, const ShapeVector xshp,
+                                                 const ShapeVector yshp) -> void {
+    output.push_back(xshp[(transpose_a ? 1 : 0)]);
+    output.push_back(yshp[(transpose_b ? 0 : 1)]);
     return;
   };
-  make_shape(ret_shape, x->shape()->shape(), y->shape()->shape());
+  make_shape(ret_shape, x_shp, y_shp);
   make_shape(ret_min_shape, x_min_shape, y_min_shape);
   make_shape(ret_max_shape, x_max_shape, y_max_shape);
-  return std::make_shared<AbstractTensor>(x->element(),
-                                          std::make_shared<Shape>(ret_shape, ret_min_shape, ret_max_shape));
+  TypePtr x_type = x->element()->GetTypeTrack();
+  if (x_type->type_id() == TypeId::kNumberTypeInt8) {
+    x_type = kInt32;
+  }
+  return std::make_shared<AbstractTensor>(x_type, std::make_shared<Shape>(ret_shape, ret_min_shape, ret_max_shape));
 }
 
 AbstractBasePtr InferImplBatchMatMul(const AnalysisEnginePtr &, const PrimitivePtr &primitive,
@@ -377,24 +395,40 @@ AbstractBasePtr InferImplBatchMatMul(const AnalysisEnginePtr &, const PrimitiveP
   auto y = CheckArg<AbstractTensor>(op_name, args_spec_list, 1);
   MS_EXCEPTION_IF_NULL(y);
   MS_EXCEPTION_IF_NULL(y->shape());
-  if (x->shape()->shape().size() != y->shape()->shape().size() || x->shape()->shape().size() < 3) {
+  auto x_shp = x->shape()->shape();
+  auto y_shp = y->shape()->shape();
+  if (x_shp.size() != y_shp.size() || x_shp.size() < 3) {
     MS_LOG(EXCEPTION)
       << "BatchMatMul input x, y should have the same dimension size and should be greater or equal to 3.";
   }
-  ValuePtr TAptr = primitive->GetAttr("transpose_a");
-  ValuePtr TBptr = primitive->GetAttr("transpose_b");
-  bool TA = GetValue<bool>(TAptr);
-  bool TB = GetValue<bool>(TBptr);
+  ValuePtr transpose_a_ptr = primitive->GetAttr("transpose_a");
+  ValuePtr transpose_b_ptr = primitive->GetAttr("transpose_b");
+  bool transpose_a = GetValue<bool>(transpose_a_ptr);
+  bool transpose_b = GetValue<bool>(transpose_b_ptr);
   ShapeVector x_min_shape = x->shape()->min_shape();
   ShapeVector x_max_shape = x->shape()->max_shape();
   ShapeVector y_min_shape = y->shape()->min_shape();
   ShapeVector y_max_shape = y->shape()->max_shape();
-  (void)CheckMinMaxShape(x->shape()->shape(), &x_min_shape, &x_max_shape);
-  (void)CheckMinMaxShape(y->shape()->shape(), &y_min_shape, &y_max_shape);
+  (void)CheckMinMaxShape(x_shp, &x_min_shape, &x_max_shape);
+  (void)CheckMinMaxShape(y_shp, &y_min_shape, &y_max_shape);
+  // Additional check for dynamic shape
+  // Last infer will be real shape values
+  bool x_not_dyn = std::all_of(x_shp.begin(), x_shp.end(), [](int64_t value) { return value != Shape::SHP_ANY; });
+  bool y_not_dyn = std::all_of(y_shp.begin(), y_shp.end(), [](int64_t value) { return value != Shape::SHP_ANY; });
+  if (x_not_dyn && y_not_dyn) {
+    size_t offset = x_shp.size() - 2;
+    auto x_col = x_shp[offset + (transpose_a ? 0 : 1)];
+    auto y_row = y_shp[offset + (transpose_b ? 1 : 0)];
+    if (x_col != y_row) {
+      MS_LOG(EXCEPTION) << "BatchMatMul shape error, got x_col: " << x_col << ", y_row: " << y_row
+                        << ". In BatchMatMul x_col and y_row should be equal.";
+    }
+  }
   ShapeVector ret_shape;
   ShapeVector ret_min_shape;
   ShapeVector ret_max_shape;
-  auto make_shape = [&TA, &TB](ShapeVector &output, const ShapeVector xshp, const ShapeVector yshp) -> void {
+  auto make_shape = [&transpose_a, &transpose_b](ShapeVector &output, const ShapeVector xshp,
+                                                 const ShapeVector yshp) -> void {
     for (size_t i = 0; i < xshp.size() - 2; i++) {
       if (xshp[i] != yshp[i]) {
         if (xshp[i] > 0 && yshp[i] > 0) {
@@ -406,15 +440,18 @@ AbstractBasePtr InferImplBatchMatMul(const AnalysisEnginePtr &, const PrimitiveP
       }
     }
     size_t offset = xshp.size() - 2;
-    output.push_back(xshp[offset + (TA ? 1 : 0)]);
-    output.push_back(yshp[offset + (TB ? 0 : 1)]);
+    output.push_back(xshp[offset + (transpose_a ? 1 : 0)]);
+    output.push_back(yshp[offset + (transpose_b ? 0 : 1)]);
     return;
   };
-  make_shape(ret_shape, x->shape()->shape(), y->shape()->shape());
+  make_shape(ret_shape, x_shp, y_shp);
   make_shape(ret_min_shape, x_min_shape, y_min_shape);
   make_shape(ret_max_shape, x_max_shape, y_max_shape);
-  return std::make_shared<AbstractTensor>(x->element(),
-                                          std::make_shared<Shape>(ret_shape, ret_min_shape, ret_max_shape));
+  TypePtr x_type = x->element()->GetTypeTrack();
+  if (x_type->type_id() == TypeId::kNumberTypeInt8) {
+    x_type = kInt32;
+  }
+  return std::make_shared<AbstractTensor>(x_type, std::make_shared<Shape>(ret_shape, ret_min_shape, ret_max_shape));
 }
 }  // namespace abstract
 }  // namespace mindspore