TEST(CAPI, TestFromProto) {
  Tensor t_cc(DT_FLOAT, TensorShape({2, 3}));
  t_cc.flat<float>().setConstant(1.0);
  tensorflow::TensorProto t_proto;
  t_cc.AsProtoField(&t_proto);

  TF_Buffer* t_buffer = TF_NewBuffer();
  TF_CHECK_OK(MessageToBuffer(t_proto, t_buffer));

  const int num_bytes = 6 * sizeof(float);
  float* values =
      reinterpret_cast<float*>(tensorflow::cpu_allocator()->AllocateRaw(

      op_->node_builder.Attr(attr_name, v);
    }
    return absl::OkStatus();
  }
  Status SetAttrFloatList(const char* attr_name, const float* values,
                          int num_values) override {
    op_->node_builder.Attr(attr_name,
                           ArraySlice<const float>(values, num_values));
    return absl::OkStatus();
  }
  Status SetAttrIntList(const char* attr_name, const int64_t* values,

        "//tensorflow/c/experimental/ops:nn_ops",
        "//tensorflow/core:test",
        "//tensorflow/core:test_main",
        "//tensorflow/core/platform:tensor_float_32_utils",
    ] + if_libtpu(
        if_false = ["//tensorflow/compiler/mlir/tensorflow/c:mlir_c_api_registration"],
        if_true = [],
    ),
)

tf_cuda_cc_test(
    name = "math_grad_test",

                     float value) {
  desc->node_builder.Attr(attr_name, value);
}

void TF_SetAttrFloatList(TF_OperationDescription* desc, const char* attr_name,
                         const float* values, int num_values) {
  desc->node_builder.Attr(attr_name,
                          ArraySlice<const float>(values, num_values));
}

          i_vector.push_back(default_value.list().i(i));
        }
        TFE_OpSetAttrIntList(op, attr_name, i_vector.data(), i_size);
      }
      // Float
      if (const int f_size = default_value.list().f_size()) {
        absl::InlinedVector<float, 4> f_vector;
        f_vector.reserve(f_size);
        for (int i = 0; i < f_size; ++i) {
          f_vector.push_back(default_value.list().f(i));
        }

        "//tensorflow/c:tf_tensor_internal",
        "//tensorflow/c/experimental/ops",
        "//tensorflow/core:test",
        "//tensorflow/core:test_main",
        "//tensorflow/core/platform:tensor_float_32_utils",
        "@com_google_absl//absl/types:span",
    ],
)

cc_library(
    name = "abstract_tensor_handle",
    srcs = ["abstract_tensor_handle.cc"],
    hdrs = ["abstract_tensor_handle.h"],

* `tf.config.experimental.enable_tensor_float_32_execution`

          tf_additional_profiler_deps()) + if_xla_available([
        "//tensorflow/compiler/aot:tfcompile_lib",
    ]) + if_static(extra_deps = [
        "//tensorflow/core/platform:tensor_float_32_utils",
        "//tensorflow/core/platform:enable_tf2_utils",
    ]) + if_oss([
        "//tensorflow/core/distributed_runtime/rpc:grpc_server_lib",
        "//tensorflow/core/distributed_runtime/rpc:grpc_session",

};

// Template instantiations here

inline bool IsDtypeTrainable(DataType dtype) {
  switch (dtype) {
    case DT_HALF:
    case DT_BFLOAT16:
    case DT_FLOAT:
    case DT_DOUBLE:
    case DT_COMPLEX64:
    case DT_COMPLEX128:
    case DT_RESOURCE:
    case DT_VARIANT:
      return true;
    case DT_QINT8:
    case DT_QINT16:
    case DT_QINT32:

          if ver < 35:
            print('WARNING: XLA does not support CUDA compute capabilities '
                  'lower than sm_35. Disable XLA when running on older GPUs.')
      else:
        ver = float(m.group(0))
        if ver < 3.0:
          print('ERROR: TensorFlow only supports CUDA compute capabilities 3.0 '
                'and higher. Please re-specify the list of compute '