// RUN: json_to_flatbuffer %p/test_schema.fbs %s | flatbuffer_translate --tflite-flatbuffer-to-mlir -o - | FileCheck %s

// CHECK: %0 = "tfl.pseudo_const"() <{value = dense<2> : tensor<2xi32>}> : () -> tensor<2xi32>
// CHECK: %1 = "tfl.reshape"(%arg0, %0) : (tensor<1x4xf32>, tensor<2xi32>) -> tensor<2x2xf32>

{
  "version": 3,
  "operator_codes": [
    {
      "builtin_code": "RESHAPE"
    }
  ],
  "subgraphs": [
    {
      "tensors": [

// RUN: tf_tfl_translate --enable-hlo-to-tf-conversion --input-mlir %s -o - | flatbuffer_translate --tflite-flatbuffer-to-mlir - -o - | FileCheck %s --check-prefix=CHECK-ROUNDTRIP


module {
  func.func public @main( %arg0: tensor<i64>) ->  tensor<i64> {
    %0 = func.call @test_add_roundtrip(%arg0) : (tensor<i64>) -> tensor<i64>

    return %0 : tensor<i64>
  }


  // CHECK-LABEL: func.func private @test_add_roundtrip

// RUN: tf-mlir-translate -mlir-tf-str-attr-to-mlir %s -mlir-print-debuginfo -mlir-print-local-scope | FileCheck %s

"\0A\0Amodule attributes {tf.versions = {producer = 888 : i32}} {\0A func.func @main(%arg0: tensor<?xi32>) -> tensor<?xi32> {\0A %0 = \22tf.Identity\22(%arg0) : (tensor<?xi32>) -> tensor<?xi32> loc(unknown)\0A return %0 : tensor<?xi32> loc(unknown)\0A } loc(unknown)\0A} loc(unknown)"

// Test simple serialized computation consisting of a function named `main`

// RUN: tf-quant-opt %s -mark-functions-noinline='noinline-functions=noinline0' \
// RUN:     -allow-unregistered-dialect -mlir-disable-threading \
// RUN:     -split-input-file -verify-diagnostics | FileCheck %s

// Tests that the function is marked tf._noinline = true.

// CHECK-LABEL: @noinline0
// CHECK-SAME: attributes {{{.*tf._noinline = true.*}}}
func.func @noinline0() -> (tensor<0xf32>) {

// RUN: tf-opt %s --tf-strip-saved-module-metadata --split-input-file | FileCheck %s

// CHECK-LABEL: module
// CHECK-NOT: tf_saved_model
module attributes {tf_saved_model.semantics} {
  // CHECK: tf_saved_model.global_tensor
  // CHECK-NOT: tf_saved_model
  "tf_saved_model.global_tensor"() { is_mutable, sym_name = "v", type = tensor<?xf32>, value = dense<1.> : tensor<1xf32> } : () -> ()

# RUN: tf-mlir-translate -graphdef-to-mlir -tf-enable-shape-inference-on-import=false %s -o - -mlir-print-debuginfo -mlir-print-local-scope | FileCheck %s

node {
  name: "Quantized_Constant"
  op: "Const"
  attr {
    key: "dtype"
    value {
      type: DT_QUINT8
    }
  }
  attr {
    key: "value"
    value {
      tensor {
        dtype: DT_QUINT8
        tensor_shape {
        }
        int_val: 7
      }
    }

// RUN: tf-tfrt-opt -tf-to-tfrt %s | FileCheck %s --dump-input=fail

// CHECK-LABEL: func @device_test
func.func @device_test(
    %arg0: tensor<3x1xf32> {tf_saved_model.index_path = [0]},
    %arg1: tensor<1x3xf32> {tf_saved_model.index_path = [0]})
      -> (tensor<3x3xf32> {tf_saved_model.index_path = []}) {
  // CHECK: {{%.*}} = corert.get_op_handler %arg0 "/device:GPU:0"

// RUN: flatbuffer_translate -mlir-to-tflite-flatbuffer %s -o - | flatbuffer_translate --tflite-flatbuffer-to-mlir - -o - | FileCheck %s
// Ensure bucketize roundtrip exactly

func.func @main(%arg0: tensor<3x2xf32>) -> tensor<3x2xi32> {
  // CHECK-LABEL: @main
  // CHECK: "tfl.bucketize"(%arg0) <{boundaries = [0.000000e+00 : f32, 1.000000e+01 : f32, 1.000000e+02 : f32]}> : (tensor<3x2xf32>) -> tensor<3x2xi32>

// RUN: flatbuffer_translate -mlir-to-tflite-flatbuffer %s -o - | flatbuffer_translate --tflite-flatbuffer-to-mlir - -o - | FileCheck %s
// Ensure cast with bfloat16 roundtrip exactly

func.func @main(tensor<4x5xbf16>) -> tensor<4x5xbf16> {
^bb0(%arg0: tensor<4x5xbf16>):
  // CHECK-LABEL: @main
  // CHECK:  (tensor<4x5xbf16>) -> tensor<4x5xf32>
  // CHECK-NEXT:  (tensor<4x5xf32>) -> tensor<4x5xbf16>

// RUN: json_to_flatbuffer %p/test_schema.fbs %s | flatbuffer_translate --tflite-flatbuffer-to-mlir -o - | FileCheck %s

// If input and output tensors don't have names, there shouldn't be an
// `tf.entry_function` attribute created.
// CHECK-NOT: tf.entry_function

{
  "version": 3,
  "operator_codes": [
    {
      "builtin_code": "CONV_2D"
    }
  ],
  "subgraphs": [
    {
      "tensors": [
        {
          "shape": [