* 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.*

        "//learning/pathways/...",
        "//learning/serving/contrib/tfrt/mlir/canonical_ops/...",
        "//learning/serving/experimental/remote_predict/...",
        "//perftools/accelerators/xprof/convert/...",
        "//perftools/accelerators/xprof/integration_tests/...",
        "//smartass/brain/...",
        "//tensorflow/...",
        "//tensorflow_decision_forests/...",
        "//tensorflow_federated/...",

    - [**Node**](#node-3)
      - [**Pod API**](#pod-api)
      - [**Hardware Accelerators**](#hardware-accelerators)
      - [**Container Runtime**](#container-runtime)
      - [**Kubelet**](#kubelet)
        - [**Other**](#other-5)
    - [**OpenStack**](#openstack-2)
    - [**Scheduling**](#scheduling-2)
      - [**Hardware Accelerators**](#hardware-accelerators-1)
      - [**Other**](#other-6)
    - [**Storage**](#storage-3)

    resource availability for critical system pods.
* **[alpha]** `--experimental-nvidia-gpus` flag is replaced by `Accelerators` alpha
  feature gate along with addition of support for multiple Nvidia GPUs.
  - To use GPUs, pass `Accelerators=true` as part of `--feature-gates` flag.
  - More information [here](https://vishh.github.io/docs/user-guide/gpus/).

by default.  The PersistentVolume NodeAffinity alpha annotation is deprecated and will be removed in a future release. ([#59391](https://github.com/kubernetes/kubernetes/pull/59391), [@msau42](https://github.com/msau42))

* The alpha Accelerators feature gate is deprecated and will be removed in v1.11. Please use device plugins ([https://github.com/kubernetes/enhancements/issues/368](https://github.com/kubernetes/enhancements/issues/368)) instead. They can be enabled using the DevicePlugins...

implementation bugs that might allow attackers to leave malicious code running
and leak or tamper with applications from other users. Please report
vulnerabilities to the vendor of the affected hardware accelerator.

## Reporting vulnerabilities

### Vulnerabilities in TensorFlow

This document covers different use cases for TensorFlow together with comments

  - [Known Issues](#known-issues)
    - [Summary API in kubelet doesn't have accelerator metrics](#summary-api-in-kubelet-doesnt-have-accelerator-metrics)
  - [Urgent Upgrade Notes](#urgent-upgrade-notes)
    - [(No, really, you MUST read this before you upgrade)](#no-really-you-must-read-this-before-you-upgrade)

- When `DisableAcceleratorUsageMetrics` is set, do not collect accelerator metrics using cAdvisor. ([#101712](https://github.com/kubernetes/kubernetes/pull/101712), [@SergeyKanzhelev](https://github.com/SergeyKanzhelev)) [SIG Instrumentation and Node]

- The kube-scheduler ComponentConfig v1beta2 is deprecated in v1.25. ([#111547](https://github.com/kubernetes/kubernetes/pull/111547), [@kerthcet](https://github.com/kerthcet))
- The kubelet no longer supports collecting accelerator metrics through cAdvisor. The feature gate `DisableAcceleratorUsageMetrics` is now GA and cannot be disabled. ([#110940](https://github.com/kubernetes/kubernetes/pull/110940), [@pacoxu](https://github.com/pacoxu))

- Adds the ability to disable Accelerator/GPU metrics collected by Kubelet ([#91930](https://github.com/kubernetes/kubernetes/pull/91930), [@RenaudWasTaken](https://github.com/RenaudWasTaken)) [SIG Node]