rRedrawOverlay();if(!preventEvent)plot.getPlaceholder().trigger("plotunselected",[])}}function extractRange(ranges,coord){var axis,from,to,key,axes=plot.getAxes();for(var k in axes){axis=axes[k];if(axis.direction==coord){key=coord+axis.n+"axis";if(!ranges[key]&&axis.n==1)key=coord+"axis";if(ranges[key]){from=ranges[key].from;to=ranges[key].to;break}}}if(!ranges[key]){axis=coord=="x"?plot.getXAxes()[0]:plot.getYAxes()[0];from=ranges[coord+"1"];to=ranges[coord+"2"]}if(from!=null&&to!=null&&from>to){var...

  StridedArrayViewBase(ArrayRef<int64_t> shape, ArrayRef<int64_t> index,
                       int64_t axis) {
    assert(shape.size() == index.size());
    assert(axis < shape.size());
    assert(axis >= 0);
    assert(index[axis] == 0);
    offset_ = IndexToOffset(shape, index);
    stride_ = StrideForAxis(shape, axis);
    size_ = shape[axis];
  }

  // Returns the size of the 1-d slice across the tensor.

      @Override
      public int size() {
        return axes.size();
      }

      @Override
      public E get(int axis) {
        checkElementIndex(axis, size());
        int axisIndex = getAxisIndexForProductIndex(index, axis);
        return axes.get(axis).get(axisIndex);
      }

      @Override
      boolean isPartialView() {
        return true;
      }

// CHECK-NEXT: %[[CC0:.*]] = "tf.RiscConcat"(%[[ED0]], %[[ED1]]) {axis = 0 : i32} : (tensor<*xf32>, tensor<*xf32>) -> tensor<*xf32>
// CHECK-NEXT: %[[ED2:.*]] = "tf.ExpandDims"(%arg2, %[[AXIS]]) : (tensor<2x3xf32>, tensor<i32>) -> tensor<*xf32>
// CHECK-NEXT: %[[CC1:.*]] = "tf.RiscConcat"(%[[CC0]], %[[ED2]]) {axis = 0 : i32} : (tensor<*xf32>, tensor<*xf32>) -> tensor<*xf32>

  xs.push_back(Placeholder(scope_, DT_FLOAT, Placeholder::Shape(shape)));
  xs.push_back(Placeholder(scope_, DT_FLOAT, Placeholder::Shape(shape)));
  auto axis = Const(scope_, 0);
  auto y = Concat(scope_, xs, axis);
  TensorShape result_shape({9, 2, 5});
  RunTest(xs, {shape, shape, shape}, {y}, {result_shape});
}

TEST_F(ArrayGradTest, BroadcastToGrad) {
  TensorShape x_shape({2, 5});

                std::vector<Output>* grad_outputs) {
  int N;
  TF_RETURN_IF_ERROR(GetNodeAttr(op.node()->attrs(), "N", &N));
  int axis;
  TF_RETURN_IF_ERROR(GetNodeAttr(op.node()->attrs(), "axis", &axis));

  grad_outputs->reserve(N);
  auto grad_op = Unstack(scope, grad_inputs[0], N, Unstack::Axis(axis));
  for (const Output& o : grad_op.output) {
    grad_outputs->emplace_back(o);
  }
  return scope.status();
}

  // CHECK-DAG: %[[ITEMS1_3:.*]] = "tf.ExpandDims"(%[[ITEMS1]]#3, %[[AXIS]])
  // CHECK-DAG: %[[ITEMS1_2:.*]] = "tf.ExpandDims"(%[[ITEMS1]]#2, %[[AXIS]])
  // CHECK-DAG: %[[ITEMS1_1:.*]] = "tf.ExpandDims"(%[[ITEMS1]]#1, %[[AXIS]])
  // CHECK-DAG: %[[ITEMS1_0:.*]] = "tf.ExpandDims"(%[[ITEMS1]]#0, %[[AXIS]])
  // CHECK-DAG: %[[ITEMS0_0:.*]] = "tf.ExpandDims"(%[[ITEMS0]], %[[AXIS]])

      @Override
      public int size() {
        return axes.size();
      }

      @Override
      public E get(int axis) {
        checkElementIndex(axis, size());
        int axisIndex = getAxisIndexForProductIndex(index, axis);
        return axes.get(axis).get(axisIndex);
      }

      @Override
      boolean isPartialView() {
        return true;
      }

}

inline bool IsUnknownDimOrRank(int64_t dim_or_rank) {
  return dim_or_rank == -1;
}

// Returns dimension index for the given TensorFlow axis that supports negative
// indexing.
inline int64_t GetDimForAxis(int64_t axis, int64_t rank) {
  return axis >= 0 ? axis : axis + rank;
}

// Returns the tf.Equal/tf.NotEqual result type given `x` and `y` and inputs. If

  kBinaryOp,        // Binary ops have lhs/rhs attr.
  kQuantizationOp,  // Quantization ops have input/output attr.
};

// For each op type, the following axis carries axis information:
// kDynamicRangeOp: rhs_quantization_axis will carry axis information.
// kUnaryOp: quantization_axis will carry axis information.
// kBinaryOp: Among {lhs, rhs, output}_quantization_axis, only check rhs.