그 다음 FastAPI와 사용하려는 다른 **패키지**를 **설치**해야 합니다.

패키지를 설치할 때는 보통 Python에 포함된 `pip` 명령어(또는 유사한 대안)를 사용합니다.

하지만 `pip`를 그대로 직접 사용하면, 패키지는 **전역 Python 환경**(전역 Python 설치)에 설치됩니다.

### 문제점 { #the-problem }

그렇다면, 전역 Python 환경에 패키지를 설치하면 어떤 문제가 있을까요?

어느 시점이 되면 **서로 다른 패키지**에 의존하는 다양한 프로그램을 작성하게 될 것입니다. 그리고 작업하는 프로젝트 중 일부는 같은 패키지의 **서로 다른 버전**에 의존할 수도 있습니다. 😱

       * `Enumeration<?>`, probably because of one of the following two bugs (and maybe those two
       * bugs are themselves just symptoms of the same underlying problem):
       *
       * https://github.com/typetools/checker-framework/issues/3030
       *
       * https://github.com/typetools/checker-framework/issues/3236
       */

       * `Enumeration<?>`, probably because of one of the following two bugs (and maybe those two
       * bugs are themselves just symptoms of the same underlying problem):
       *
       * https://github.com/typetools/checker-framework/issues/3030
       *
       * https://github.com/typetools/checker-framework/issues/3236
       */

- Fixed the bug when swap related metrics were not available in `/metrics/resource` endpoint. ([#132065](https://github.com/kubernetes/kubernetes/pull/132065), [@yuanwang04](https://github.com/yuanwang04)) [SIG Node and Testing]

- Kubeadm: remove the "ephemeral_storage" request from the etcd static pod that kubeadm deploys on stacked etcd control plane nodes. This request has caused sporadic failures on some setups due to a problem in the kubelet with cadvisor and the LocalStorageCapacityIsolation feature gate. See this issue for more details: https://github.com/kubernetes/kubernetes/issues/99305 ([#102673](https://github.com/kubernetes/kubernetes/pull/102673), [@jackfranci...

		// finally convert it back to the server type (again via xml).
		// This is needed as there is no add-rule function in the server
		// yet.

		// Though we do not check if the rule already exists, this is
		// not a problem as we are always using the same replication
		// rule ID - if the rule already exists, it is just replaced.
		replicationConfigS, _, err := globalBucketMetadataSys.GetReplicationConfig(ctx, bucket)
		if err != nil {

## Breaking Changes

*   `tf.keras`:
    *   In `tf.keras.applications` the name of the "top" layer has been
        standardized to "predictions". This is only a problem if your code
        relies on the exact name of the layer.
    *   Huber loss function has been updated to be consistent with other Keras
        losses. It now computes mean over the last axis of per-sample losses

    * [fluentd-gcp addon] Bump prometheus-to-sd to v0.5.0 to pick up security fixes.
    * [metatada-proxy addon] Bump prometheus-to-sd v0.5.0 to pick up security fixes.
* Node-Problem-Detector configuration is now decoupled from the Kubernetes release on GKE/GCE. ([#73288](https://github.com/kubernetes/kubernetes/pull/73288), [@wangzhen127](https://github.com/wangzhen127))

* Fixes a NPD bug on GCI, so that it disables glog writing to files for log-counter ([#76211](https://github.com/kubernetes/kubernetes/pull/76211), [@wangzhen127](https://github.com/wangzhen127))
* Node-Problem-Detector configuration is now decoupled from the Kubernetes release on GKE/GCE. ([#73288](https://github.com/kubernetes/kubernetes/pull/73288), [@wangzhen127](https://github.com/wangzhen127))

- Fix problem in updating VolumeAttached in node status ([#112303](https://github.com/kubernetes/kubernetes/pull/112303), [@xing-yang](https://github.com/xing-yang)) [SIG Apps]