# ARM64 WHEEL
test:linux_arm64_wheel_test_filters --test_tag_filters=-no_oss,-tf_tosa,-no_aarch64,-oss_excluded,-oss_serial,-gpu,-tpu,-benchmark-test,-v1only,-no_oss_py310,-no_oss_py313
test:linux_arm64_wheel_test_filters --build_tag_filters=-no_oss,-tf_tosa,-no_aarch64,-oss_excluded,-oss_serial,-gpu,-tpu,-benchmark-test,-v1only,-no_oss_py310,-no_oss_py313

         * It has an added advantage since you can you easily switch to different hardware accelerators (cpu, gpu, tpu) as per the task.
         * All you need is a good internet connection and you are all set.
      * Try to build TF from sources by changing CPU optimization flags.
   
   *Please let us know if this helps.*
   

              TFE_TensorHandle* axis);

// If there is a device of type `device_type`, returns true
// and sets 'device_name' accordingly.
// `device_type` must be either "GPU" or "TPU".
bool GetDeviceName(TFE_Context* ctx, tensorflow::string* device_name,
                   const char* device_type);

// Create a ServerDef with the given `job_name` and add `num_tasks` tasks in it.

config_setting(
    name = "disable_mlir_bridge",
    define_values = {"enable_mlir_bridge": "false"},
    visibility = ["//visibility:public"],
)

# This flag enables experimental TPU support
bool_flag(
    name = "enable_tpu_support",
    build_setting_default = False,
)

config_setting(
    name = "with_tpu_support_define",
    define_values = {"with_tpu_support": "true"},

*   `tf.tpu.experimental.embedding`:

    *   `tf.tpu.experimental.embedding.FeatureConfig` now takes an additional
        argument `output_shape` which can specify the shape of the output
        activation for the feature.
    *   `tf.tpu.experimental.embedding.TPUEmbedding` now has the same behavior
        as `tf.tpu.experimental.embedding.serving_embedding_lookup` which can

* Fix regression in `v1.JobSpec.backoffLimit` that caused failed Jobs to be restarted indefinitely. ([#63650](https://github.com/kubernetes/kubernetes/pull/63650), [@soltysh](https://github.com/soltysh))
* GCE: Update cloud provider to use TPU v1 API ([#64727](https://github.com/kubernetes/kubernetes/pull/64727), [@yguo0905](https://github.com/yguo0905))

- cloud.google.com/go/speech: v1.21.0
- cloud.google.com/go/storagetransfer: v1.10.3
- cloud.google.com/go/talent: v1.6.5
- cloud.google.com/go/texttospeech: v1.7.4
- cloud.google.com/go/tpu: v1.6.4
- cloud.google.com/go/trace: v1.10.4
- cloud.google.com/go/translate: v1.10.0
- cloud.google.com/go/video: v1.20.3
- cloud.google.com/go/videointelligence: v1.11.4
- cloud.google.com/go/vision/v2: v2.7.5

* GCE: Update cloud provider to use TPU v1 API ([#64727](https://github.com/kubernetes/kubernetes/pull/64727), [@yguo0905](https://github.com/yguo0905))

- cloud.google.com/go/speech: v1.15.0 → v1.19.0
- cloud.google.com/go/storagetransfer: v1.8.0 → v1.10.0
- cloud.google.com/go/talent: v1.5.0 → v1.6.2
- cloud.google.com/go/texttospeech: v1.6.0 → v1.7.1
- cloud.google.com/go/tpu: v1.5.0 → v1.6.1
- cloud.google.com/go/trace: v1.9.0 → v1.10.1
- cloud.google.com/go/translate: v1.7.0 → v1.8.2
- cloud.google.com/go/video: v1.15.0 → v1.19.0
- cloud.google.com/go/videointelligence: v1.10.0 → v1.11.1

- cloud.google.com/go/speech: v1.15.0
- cloud.google.com/go/storagetransfer: v1.8.0
- cloud.google.com/go/talent: v1.5.0
- cloud.google.com/go/texttospeech: v1.6.0
- cloud.google.com/go/tpu: v1.5.0
- cloud.google.com/go/trace: v1.9.0
- cloud.google.com/go/translate: v1.7.0
- cloud.google.com/go/video: v1.15.0
- cloud.google.com/go/videointelligence: v1.10.0
- cloud.google.com/go/vision/v2: v2.7.0