// RUN: flatbuffer_translate -mlir-to-tflite-flatbuffer %s -o - | flatbuffer_to_string - | FileCheck %s

func.func @main(tensor<40x37xf32>, tensor<40x37xf32>) -> tensor<40x40xf32> {
^bb0(%arg0: tensor<40x37xf32>, %arg1: tensor<40x37xf32>):
  // CHECK:      {
  // CHECK-NEXT:  version: 3,
  // CHECK-NEXT:  operator_codes: [ {
  // CHECK-NEXT:    deprecated_builtin_code: 9,
  // CHECK-NEXT:    version: 2,
  // CHECK-NEXT:    builtin_code: FULLY_CONNECTED

    func.return %val : tensor<f32>
  }

  // CHECK: func private @f_callee_callee(%arg0: tensor<*x!tf_type.resource>) -> tensor<f32>
  func.func private @f_callee_callee(%arg0: tensor<*x!tf_type.resource>) -> tensor<f32> {
    %c0 = "tf.Const"() { value = dense<1.0> : tensor<f32> } : () -> tensor<f32>
    "tf.AssignVariableOp"(%arg0, %c0) : (tensor<*x!tf_type.resource>, tensor<f32>) -> ()

    %cst = arith.constant dense<0.000000e+00> : tensor<5xf32>
    %cst_0 = arith.constant dense<1.0> : tensor<5x3xf32>
    %0 = "tfl.fully_connected"(%arg0, %cst_0, %cst) {fused_activation_function = "NONE", keep_num_dims = false, weights_format = "DEFAULT"} : (tensor<?x3xf32>, tensor<5x3xf32>, tensor<5xf32>) -> tensor<?x5xf32>
    func.return %0, %0 : tensor<?x5xf32>, tensor<?x5xf32>
  }

  %0 = "tfl.pseudo_const" () {value = dense<[[1, 2], [3, 4], [5, 6]]> : tensor<3x2xi32>} : () -> tensor<3x2xi32> loc("Const")
  %1,%control_1 = tfl.control_node controls "tfl.sub" (%arg0, %0) {fused_activation_function = "RELU6"} : (tensor<3x2xi32>, tensor<3x2xi32>) -> tensor<3x2xi32> loc("sub")
  %2 = "arith.constant" () {value = dense<10> : tensor<i32>} : () -> tensor<i32> loc("SameNameAsOutput")

  %0 = stablehlo.dot_general %arg0, %arg1, contracting_dims = [1] x [0] : (tensor<1x1024xf32>, tensor<1024x3xf32>) -> tensor<1x3xf32>
  return %0 : tensor<1x3xf32>
}
// CHECK-LABEL: func.func @main
// CHECK-SAME: (%[[ARG_0:.+]]: tensor<1x1024xf32>) -> tensor<1x3xf32>

// Tests that the dot_general accepts i8 tensors and outputs an i32 tensor.

// just returns the element type of its first tensor, which is only meaningful
// when the variadic operand has at least one tensor and the tensors all have
// the same element type.
class TF_DerivedOperandTypeAttr<int idx> : DerivedTypeAttr<
  "return mlir::getElementTypeOrSelf(*getODSOperands(" # idx # ").begin());">;

// A derived attribute that returns the element types of the tensors in the

};

inline Tensor TensorHandle::Resolve(Status* status) {
  TF_Tensor* tensor =
      TFE_TensorHandleResolve(handle_.get(), status->GetTFStatus());
  if (!status->ok()) {
    return Tensor(nullptr);
  }
  return Tensor(tensor);
}

inline TensorHandle TensorHandle::FromTensor(const Tensor& tensor,
                                             const Runtime& runtime,

  // CHECK-NEXT: }

  %0 = "tfl.pseudo_const" () {value = dense<1.0> : tensor<4xf32>} : () -> tensor<4xf32> loc("Const")
  %1 = "tfl.squared_difference"(%arg0, %0) {fused_activation_function = "NONE"} : (tensor<4xf32>, tensor<4xf32>) -> tensor<4xf32> loc("squared_difference")
  %2 = "tfl.mul"(%arg0, %1) {fused_activation_function = "NONE"} : (tensor<4xf32>, tensor<4xf32>) -> tensor<4xf32> loc("mul")

    }

    // Extract individual tensor list element and combine them using the tf.Pack
    // op.
    Location loc = op.getLoc();
    llvm::SmallVector<Value, 4> values;
    values.reserve(tensors.size());
    for (const tensorflow::Tensor &tensor : tensors) {
      auto attr_or = tensorflow::ConvertTensor(tensor, &rewriter);
      if (!attr_or.ok()) return failure();

  %0 = "tfl.pseudo_const" () {value = dense<-1.23697901> : tensor<32xf32>} : () -> tensor<32xf32> loc("Const")
  %1 = "tfl.pseudo_qconst"() {qtype = tensor<32x3x3x3x!quant.uniform<u8<1:255>:f32, 0.021826678373682216:151>>, value = dense<-76> : tensor<32x3x3x3xi8>} : () -> tensor<32x3x3x3x!quant.uniform<u8<1:255>:f32, 0.021826678373682216:151>>