## 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))

* Kubelet now exposes a new endpoint, `/metrics/probes`, which exposes a Prometheus metric containing the liveness and/or readiness probe results for a container. ([#61369](https://github.com/kubernetes/kubernetes/pull/61369), [@rramkumar1](https://github.com/rramkumar1))

### SIG Network

- Fix: ignore non-VMSS error for VMAS nodes in `reconcileBackendPools`. ([#103997](https://github.com/kubernetes/kubernetes/pull/103997), [@nilo19](https://github.com/nilo19))
- Fix: leave the probe path empty for TCP probes ([#105253](https://github.com/kubernetes/kubernetes/pull/105253), [@nilo19](https://github.com/nilo19))

back and look at this code closely (wtf?)\n  // If the content of the popper grows once it's been positioned, it\n  // may happen that the popper gets misplaced because of the new content\n  // overflowing its reference element\n  // To avoid this problem, we provide two options (x and y), which allow\n  // the consumer to define the offset origin.\n  // If we position a popper on top of a reference element, we can set\n  // `x` to `top` to make the popper grow towards its top instead of\n  // its...

    * This can solve nuanced errors when using middlewares. Before Starlette `0.24.0`, a new instance of each middleware class would be created when a new middleware was added. That normally was not a problem, unless the middleware class expected to be created only once, with only one instance, that happened in some cases. This upgrade would solve those cases (thanks [@adriangb](https://github.com/adriangb)! Starlette PR [#2017](https://github.com/enc...

                validateStringNotEmpty(problems, "mirrors.mirror.id", mirror.getId(), mirror.getUrl());

                validateBannedCharacters(
                        problems, "mirrors.mirror.id", Severity.WARNING, mirror.getId(), null, ILLEGAL_REPO_ID_CHARS);

                if ("local".equals(mirror.getId())) {
                    addViolation(
                            problems,
                            Severity.WARNING,

        interpolateObject(model, model, projectDir, config, problems);
        return model;
    }

    void interpolateObject(
            Object obj, Model model, File projectDir, ModelBuildingRequest config, ModelProblemCollector problems) {
        List<? extends ValueSource> valueSources = createValueSources(model, projectDir, config, problems);

    private final ModelBuildingRequest request;

    private final ModelProblemCollector problems;

    DefaultModelBuildingEvent(Model model, ModelBuildingRequest request, ModelProblemCollector problems) {
        this.model = model;
        this.request = request;
        this.problems = problems;
    }

    @Override
    public Model getModel() {
        return model;
    }

    @Override

        repo.setUrl("http://xxx.xxx.com");
        problems = new SimpleProblemCollector();
        validator.validate(model, problems);
        assertEquals(1, problems.messages.size());

        repo.setId("xxx");
        problems = new SimpleProblemCollector();
        validator.validate(model, problems);
        assertEquals(0, problems.messages.size());
    }

    @Test
    void testValidateMirror() throws Exception {