Yingding Wang

I assume this means the `app:initialColor` is not set, thus the default [initalColor](https://github.com/QuadFlask/colorpicker/blob/master/library/src/main/res/values/attrs.xml) `White` is chosen. I don't know how to deactivate the initialColor but you can use a different...

It is exactly what I need to upgrade from KF 1.4.0 to KF 1.5.0 with manifests: * I was not sure whether I shall leave out the KF 1.4.1. *...

@andyzhangx Thanks for your feedback. > if `csi-nfs-controller` is running on one worker node which is restarting, the `csi-nfs-controller` would not accept any request, so that's by design. But what...

@andyzhangx Thanks, next time wenn i run into this issue. I will try `kubectl logs csi-nfs-controller-56bfddd689-dh5tk -c nfs -n kube-system -c csi-provisioner` to see if `-c csi-provisioner` option gives me...

@andyzhangx The same issue happens again. From `kubectl logs csi-nfs-controller-656d5f9d5b-9xg9p -c nfs -n kube-system -c csi-provisioner` I got the following error output. ``` ... I0628 10:24:41.136751 1 controller.go:1337] provision "kubeflow/test-volume"...

ok. I fall back to helm3 chart v4.0.0 and every thing works now. It is an issue of v4.1.0.

Thanks for clarifying, I thought initially I have to use v4.1.0 since it supports 1.20+ and my k8s is 1.21.12

In a pyhton3.8 virtual env, for the tf base version 2.5 and tensorflow_macos==2.5.0 and tensorflow-metal seem to work: ``` SYSTEM_VERSION_COMPAT=0 pip install tensorflow-macos==2.5.0 tensorflow-federated==0.19.0 tensorflow-metal ```

I tested the tff example from https://www.tensorflow.org/federated, it is kind of weird to see the cpu is faster. But why? ``` | type | time in secs | |-----------|--------------| |...

An [update](https://github.com/tensorflow/federated/issues/2307#issuecomment-1073244816) on this with an intel based MBP works with [email protected] ``` SYSTEM_VERSION_COMPAT=0 pip install tensorflow-macos==2.8.0 tensorflow-metal==0.4.0; SYSTEM_VERSION_COMPAT=0 pip install tensorflow-federated==0.20.0 ```