$ sudo -s
# curl -s https://packages.cloud.google.com/apt/doc/apt-key.gpg | apt    -key add -
OK
# cat <<EOF >/etc/apt/sources.list.d/kubernetes.list
> deb https://apt.kubernetes.io/ kubernetes-xenial main
> EOF
# apt-get update
# apt-get install -y kubelet kubeadm kubectl
#  kubeadm version -o yaml
clientVersion:
  buildDate: "2021-06-16T12:57:56Z"
  compiler: gc
  gitCommit: 092fbfbf53427de67cac1e9fa54aaa09a28371d7
  gitTreeState: clean
  gitVersion: v1.21.2
  goVersion: go1.16.5
  major: "1"
  minor: "21"
  platform: linux/arm64
#  cat /proc/sys/net/bridge/bridge-nf-call-iptables
1
$ kubeadm version
kubeadm version: &version.Info{Major:"1", Minor:"21", GitVersion:"v1.21.2", GitC    ommit:"092fbfbf53427de67cac1e9fa54aaa09a28371d7", GitTreeState:"clean", BuildDat    e:"2021-06-16T12:57:56Z", GoVersion:"go1.16.5", Compiler:"gc", Platform:"linux/a    rm64"}
# swapoff -a
# kubeadm init --pod-network-cidr=10.244.0.0/16
[init] Using Kubernetes version: v1.21.2
{中略}
Your Kubernetes control-plane has initialized successfully!

To start using your cluster, you need to run the following as a regular user:

  mkdir -p $HOME/.kube
  sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
  sudo chown $(id -u):$(id -g) $HOME/.kube/config

Alternatively, if you are the root user, you can run:

  export KUBECONFIG=/etc/kubernetes/admin.conf

You should now deploy a pod network to the cluster.
Run "kubectl apply -f [podnetwork].yaml" with one of the options listed at:
  https://kubernetes.io/docs/concepts/cluster-administration/addons/

Then you can join any number of worker nodes by running the following on each as root:

kubeadm join 192.168.68.111:6443 --token iuzu6k.j2arujghto188qq1 \
        --discovery-token-ca-cert-hash sha256:bed560334a382d997a48491083e569dbaaac8b1a6d8804c9b917b8596d36b255

$ sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
$ sudo chown $(id -u):$(id -g) $HOME/.kube/config

$  kubectl get node
The connection to the server 127.0.0.1:16443 was refused - did you specify the right host or port?

$ helm search hub prometheus　のコマンドは;
$ microk8s.helm3 search hub prometheus のように読み替え

$ micro8ks.helm3 repo add stable https://charts.helm.sh/stable
$ microk8s.helm3 repo add bitnami https://charts.bitnami.com/bitnami

$ microk8s helm3 repo list
WARNING: Kubernetes configuration file is group-readable. This is insecure. Location: /var/snap/microk8s/2265/credentials/client.config
NAME    URL
stable  https://charts.helm.sh/stable
bitnami https://charts.bitnami.com/bitnami

$ microk8s helm3 search repo wordpress
WARNING: Kubernetes configuration file is group-readable. This is insecure. Location: /var/snap/microk8s/2265/credentials/client.config
NAME                    CHART VERSION   APP VERSION     DESCRIPTION
bitnami/wordpress       11.0.16         5.7.2           Web publishing platform for building blogs and ...
stable/wordpress        9.0.3           5.3.2           DEPRECATED Web publishing platform for building...

# namespaceを作成
$ kubectl create namespace helm-test
# dry-run
$ helm install test stable/prometheus --namespace helm-test --dry-run
# デプロイ
$ helm install stable/prometheus --name test --namespace helm-test
# 確認
$ helm list -n helm-test
$ kubectl get po -n helm-test

$ microk8s helm3 install test bitnami/wordpress --namespace helm-test --dry-run
WARNING: Kubernetes configuration file is group-readable. This is insecure. Location: /var/snap/microk8s/2265/credentials/client.config
NAME: test
LAST DEPLOYED: Tue Jun 22 10:03:29 2021
NAMESPACE: helm-test
STATUS: pending-install
REVISION: 1
HOOKS:
---
# Source: wordpress/templates/tests/test-mariadb-connection.yaml
apiVersion: v1
kind: Pod
metadata:
  name: "test-credentials-test"
  annotations:
    "helm.sh/hook": test-success
spec:
  securityContext:
    fsGroup: 1001
  containers:
    - name: test-credentials-test
      image: docker.io/bitnami/wordpress:5.7.2-debian-10-r25
      imagePullPolicy: "IfNotPresent"
      securityContext:
        runAsNonRoot: true
        runAsUser: 1001
      env:
        - name: MARIADB_HOST
          value: "test-mariadb"
        - name: MARIADB_PORT
          value: "3306"
        - name: WORDPRESS_DATABASE_NAME
          value: "bitnami_wordpress"
        - name: WORDPRESS_DATABASE_USER
          value: "bn_wordpress"
        - name: WORDPRESS_DATABASE_PASSWORD
          valueFrom:
            secretKeyRef:
              name: test-mariadb
              key: mariadb-password
      command:
        - /bin/bash
        - -ec
        - |
          mysql --host=$MARIADB_HOST --port=$MARIADB_PORT --user=$WORDPRESS_DATABASE_USER --password=$WORDPRESS_DATABASE_PASSWORD
  restartPolicy: Never
MANIFEST:
---
# Source: wordpress/charts/mariadb/templates/serviceaccount.yaml
apiVersion: v1
kind: ServiceAccount
metadata:
  name: test-mariadb
  namespace: helm-test
  labels:
    app.kubernetes.io/name: mariadb
    helm.sh/chart: mariadb-9.3.14
    app.kubernetes.io/instance: test
    app.kubernetes.io/managed-by: Helm
  annotations:
---
# Source: wordpress/charts/mariadb/templates/secrets.yaml
apiVersion: v1
kind: Secret
metadata:
  name: test-mariadb
  namespace: helm-test
  labels:
    app.kubernetes.io/name: mariadb
    helm.sh/chart: mariadb-9.3.14
    app.kubernetes.io/instance: test
    app.kubernetes.io/managed-by: Helm
type: Opaque
data:
  mariadb-root-password: "eGMyb0NNWXZVUg=="
  mariadb-password: "SFc1WlkwNWpsMw=="
---
# Source: wordpress/templates/secrets.yaml
apiVersion: v1
kind: Secret
metadata:
  name: test-wordpress
  namespace: "helm-test"
  labels:
    app.kubernetes.io/name: wordpress
    helm.sh/chart: wordpress-11.0.16
    app.kubernetes.io/instance: test
    app.kubernetes.io/managed-by: Helm
type: Opaque
data:
  wordpress-password: "TVJteHlENlFtQQ=="
---
# Source: wordpress/charts/mariadb/templates/primary/configmap.yaml
apiVersion: v1
kind: ConfigMap
metadata:
  name: test-mariadb
  namespace: helm-test
  labels:
    app.kubernetes.io/name: mariadb
    helm.sh/chart: mariadb-9.3.14
    app.kubernetes.io/instance: test
    app.kubernetes.io/managed-by: Helm
    app.kubernetes.io/component: primary
data:
  my.cnf: |-
    [mysqld]
    skip-name-resolve
    explicit_defaults_for_timestamp
    basedir=/opt/bitnami/mariadb
    plugin_dir=/opt/bitnami/mariadb/plugin
    port=3306
    socket=/opt/bitnami/mariadb/tmp/mysql.sock
    tmpdir=/opt/bitnami/mariadb/tmp
    max_allowed_packet=16M
    bind-address=0.0.0.0
    pid-file=/opt/bitnami/mariadb/tmp/mysqld.pid
    log-error=/opt/bitnami/mariadb/logs/mysqld.log
    character-set-server=UTF8
    collation-server=utf8_general_ci

    [client]
    port=3306
    socket=/opt/bitnami/mariadb/tmp/mysql.sock
    default-character-set=UTF8
    plugin_dir=/opt/bitnami/mariadb/plugin

    [manager]
    port=3306
    socket=/opt/bitnami/mariadb/tmp/mysql.sock
    pid-file=/opt/bitnami/mariadb/tmp/mysqld.pid
---
# Source: wordpress/templates/pvc.yaml
kind: PersistentVolumeClaim
apiVersion: v1
metadata:
  name: test-wordpress
  namespace: "helm-test"
  labels:
    app.kubernetes.io/name: wordpress
    helm.sh/chart: wordpress-11.0.16
    app.kubernetes.io/instance: test
    app.kubernetes.io/managed-by: Helm
spec:
  accessModes:
    - "ReadWriteOnce"
  resources:
    requests:
      storage: "10Gi"
---
# Source: wordpress/charts/mariadb/templates/primary/svc.yaml
apiVersion: v1
kind: Service
metadata:
  name: test-mariadb
  namespace: helm-test
  labels:
    app.kubernetes.io/name: mariadb
    helm.sh/chart: mariadb-9.3.14
    app.kubernetes.io/instance: test
    app.kubernetes.io/managed-by: Helm
    app.kubernetes.io/component: primary
  annotations:
spec:
  type: ClusterIP
  ports:
    - name: mysql
      port: 3306
      protocol: TCP
      targetPort: mysql
      nodePort: null
  selector:
    app.kubernetes.io/name: mariadb
    app.kubernetes.io/instance: test
    app.kubernetes.io/component: primary
---
# Source: wordpress/templates/svc.yaml
apiVersion: v1
kind: Service
metadata:
  name: test-wordpress
  namespace: "helm-test"
  labels:
    app.kubernetes.io/name: wordpress
    helm.sh/chart: wordpress-11.0.16
    app.kubernetes.io/instance: test
    app.kubernetes.io/managed-by: Helm
spec:
  type: LoadBalancer
  externalTrafficPolicy: "Cluster"
  ports:
    - name: http
      port: 80
      protocol: TCP
      targetPort: http
    - name: https
      port: 443
      protocol: TCP
      targetPort: https
  selector:
    app.kubernetes.io/name: wordpress
    app.kubernetes.io/instance: test
---
# Source: wordpress/templates/deployment.yaml
apiVersion: apps/v1
kind: Deployment
metadata:
  name: test-wordpress
  namespace: "helm-test"
  labels:
    app.kubernetes.io/name: wordpress
    helm.sh/chart: wordpress-11.0.16
    app.kubernetes.io/instance: test
    app.kubernetes.io/managed-by: Helm
spec:
  selector:
    matchLabels:
      app.kubernetes.io/name: wordpress
      app.kubernetes.io/instance: test
  strategy:
    rollingUpdate: {}
    type: RollingUpdate
  replicas: 1
  template:
    metadata:
      labels:
        app.kubernetes.io/name: wordpress
        helm.sh/chart: wordpress-11.0.16
        app.kubernetes.io/instance: test
        app.kubernetes.io/managed-by: Helm
    spec:

      serviceAccountName: default
      # yamllint disable rule:indentation
      hostAliases:
        - hostnames:
          - status.localhost
          ip: 127.0.0.1
      # yamllint enable rule:indentation
      affinity:
        podAffinity:

        podAntiAffinity:
          preferredDuringSchedulingIgnoredDuringExecution:
            - podAffinityTerm:
                labelSelector:
                  matchLabels:
                    app.kubernetes.io/name: wordpress
                    app.kubernetes.io/instance: test
                namespaces:
                  - "helm-test"
                topologyKey: kubernetes.io/hostname
              weight: 1
        nodeAffinity:

      securityContext:
        fsGroup: 1001
      containers:
        - name: wordpress
          image: docker.io/bitnami/wordpress:5.7.2-debian-10-r25
          imagePullPolicy: "IfNotPresent"
          securityContext:
            runAsNonRoot: true
            runAsUser: 1001
          env:
            - name: ALLOW_EMPTY_PASSWORD
              value: "yes"
            - name: MARIADB_HOST
              value: "test-mariadb"
            - name: MARIADB_PORT_NUMBER
              value: "3306"
            - name: WORDPRESS_DATABASE_NAME
              value: "bitnami_wordpress"
            - name: WORDPRESS_DATABASE_USER
              value: "bn_wordpress"
            - name: WORDPRESS_DATABASE_PASSWORD
              valueFrom:
                secretKeyRef:
                  name: test-mariadb
                  key: mariadb-password
            - name: WORDPRESS_USERNAME
              value: "user"
            - name: WORDPRESS_PASSWORD
              valueFrom:
                secretKeyRef:
                  name: test-wordpress
                  key: wordpress-password
            - name: WORDPRESS_EMAIL
              value: "user@example.com"
            - name: WORDPRESS_FIRST_NAME
              value: "FirstName"
            - name: WORDPRESS_LAST_NAME
              value: "LastName"
            - name: WORDPRESS_HTACCESS_OVERRIDE_NONE
              value: "no"
            - name: WORDPRESS_ENABLE_HTACCESS_PERSISTENCE
              value: "no"
            - name: WORDPRESS_BLOG_NAME
              value: "User's Blog!"
            - name: WORDPRESS_SKIP_BOOTSTRAP
              value: "no"
            - name: WORDPRESS_TABLE_PREFIX
              value: "wp_"
            - name: WORDPRESS_SCHEME
              value: "http"
            - name: WORDPRESS_EXTRA_WP_CONFIG_CONTENT
              value:
            - name: WORDPRESS_AUTO_UPDATE_LEVEL
              value: "none"
            - name: WORDPRESS_PLUGINS
              value: "none"
          envFrom:
          ports:
            - name: http
              containerPort: 8080
            - name: https
              containerPort: 8443
          livenessProbe:
            failureThreshold: 6
            httpGet:
              httpHeaders: []
              path: /wp-admin/install.php
              port: http
              scheme: HTTP
            initialDelaySeconds: 120
            periodSeconds: 10
            successThreshold: 1
            timeoutSeconds: 5
          readinessProbe:
            failureThreshold: 6
            httpGet:
              httpHeaders: []
              path: /wp-login.php
              port: http
              scheme: HTTP
            initialDelaySeconds: 30
            periodSeconds: 10
            successThreshold: 1
            timeoutSeconds: 5
          resources:
            limits: {}
            requests:
              cpu: 300m
              memory: 512Mi
          volumeMounts:
            - mountPath: /bitnami/wordpress
              name: wordpress-data
              subPath: wordpress
      volumes:
        - name: wordpress-data
          persistentVolumeClaim:
            claimName: test-wordpress
---
# Source: wordpress/charts/mariadb/templates/primary/statefulset.yaml
apiVersion: apps/v1
kind: StatefulSet
metadata:
  name: test-mariadb
  namespace: helm-test
  labels:
    app.kubernetes.io/name: mariadb
    helm.sh/chart: mariadb-9.3.14
    app.kubernetes.io/instance: test
    app.kubernetes.io/managed-by: Helm
    app.kubernetes.io/component: primary
spec:
  replicas: 1
  revisionHistoryLimit: 10
  selector:
    matchLabels:
      app.kubernetes.io/name: mariadb
      app.kubernetes.io/instance: test
      app.kubernetes.io/component: primary
  serviceName: test-mariadb
  updateStrategy:
    type: RollingUpdate
  template:
    metadata:
      annotations:
        checksum/configuration: ba8296f4257f44a12c500b7f1720b6f3c44eb6b885a21e83bc3175cf4859939f
      labels:
        app.kubernetes.io/name: mariadb
        helm.sh/chart: mariadb-9.3.14
        app.kubernetes.io/instance: test
        app.kubernetes.io/managed-by: Helm
        app.kubernetes.io/component: primary
    spec:

      serviceAccountName: test-mariadb
      affinity:
        podAffinity:

        podAntiAffinity:
          preferredDuringSchedulingIgnoredDuringExecution:
            - podAffinityTerm:
                labelSelector:
                  matchLabels:
                    app.kubernetes.io/name: mariadb
                    app.kubernetes.io/instance: test
                    app.kubernetes.io/component: primary
                namespaces:
                  - "helm-test"
                topologyKey: kubernetes.io/hostname
              weight: 1
        nodeAffinity:

      securityContext:
        fsGroup: 1001
      containers:
        - name: mariadb
          image: docker.io/bitnami/mariadb:10.5.10-debian-10-r18
          imagePullPolicy: "IfNotPresent"
          securityContext:
            runAsUser: 1001
          env:
            - name: BITNAMI_DEBUG
              value: "false"
            - name: MARIADB_ROOT_PASSWORD
              valueFrom:
                secretKeyRef:
                  name: test-mariadb
                  key: mariadb-root-password
            - name: MARIADB_USER
              value: "bn_wordpress"
            - name: MARIADB_PASSWORD
              valueFrom:
                secretKeyRef:
                  name: test-mariadb
                  key: mariadb-password
            - name: MARIADB_DATABASE
              value: "bitnami_wordpress"
          ports:
            - name: mysql
              containerPort: 3306
          livenessProbe:
            failureThreshold: 3
            initialDelaySeconds: 120
            periodSeconds: 10
            successThreshold: 1
            timeoutSeconds: 1
            exec:
              command:
                - /bin/bash
                - -ec
                - |
                  password_aux="${MARIADB_ROOT_PASSWORD:-}"
                  if [[ -f "${MARIADB_ROOT_PASSWORD_FILE:-}" ]]; then
                      password_aux=$(cat "$MARIADB_ROOT_PASSWORD_FILE")
                  fi
                  mysqladmin status -uroot -p"${password_aux}"
          readinessProbe:
            failureThreshold: 3
            initialDelaySeconds: 30
            periodSeconds: 10
            successThreshold: 1
            timeoutSeconds: 1
            exec:
              command:
                - /bin/bash
                - -ec
                - |
                  password_aux="${MARIADB_ROOT_PASSWORD:-}"
                  if [[ -f "${MARIADB_ROOT_PASSWORD_FILE:-}" ]]; then
                      password_aux=$(cat "$MARIADB_ROOT_PASSWORD_FILE")
                  fi
                  mysqladmin status -uroot -p"${password_aux}"
          resources:
            limits: {}
            requests: {}
          volumeMounts:
            - name: data
              mountPath: /bitnami/mariadb
            - name: config
              mountPath: /opt/bitnami/mariadb/conf/my.cnf
              subPath: my.cnf
      volumes:
        - name: config
          configMap:
            name: test-mariadb
  volumeClaimTemplates:
    - metadata:
        name: data
        labels:
          app.kubernetes.io/name: mariadb
          app.kubernetes.io/instance: test
          app.kubernetes.io/component: primary
      spec:
        accessModes:
          - "ReadWriteOnce"
        resources:
          requests:
            storage: "8Gi"

NOTES:
** Please be patient while the chart is being deployed **

Your WordPress site can be accessed through the following DNS name from within your cluster:

    test-wordpress.helm-test.svc.cluster.local (port 80)

To access your WordPress site from outside the cluster follow the steps below:

1. Get the WordPress URL by running these commands:

  NOTE: It may take a few minutes for the LoadBalancer IP to be available.
        Watch the status with: 'kubectl get svc --namespace helm-test -w test-wordpress'

   export SERVICE_IP=$(kubectl get svc --namespace helm-test test-wordpress --template "{{ range (index .status.loadBalancer.ingress 0) }}{{.}}{{ end }}")
   echo "WordPress URL: http://$SERVICE_IP/"
   echo "WordPress Admin URL: http://$SERVICE_IP/admin"

2. Open a browser and access WordPress using the obtained URL.

3. Login with the following credentials below to see your blog:

  echo Username: user
  echo Password: $(kubectl get secret --namespace helm-test test-wordpress -o jsonpath="{.data.wordpress-password}" | base64 --decode)

 microk8s enable storage

Enabling default storage class
[sudo] mars のパスワード: 

deployment.apps/hostpath-provisioner created
storageclass.storage.k8s.io/microk8s-hostpath created
serviceaccount/microk8s-hostpath created
clusterrole.rbac.authorization.k8s.io/microk8s-hostpath created
clusterrolebinding.rbac.authorization.k8s.io/microk8s-hostpath created
Storage will be available soon

sudo snap install kontena-lens --classic

kontena-lens 

info: 📟 Setting Lens as protocol client for lens://
info: 📟 failed ❗
info: 🚀 Starting Lens from "/home/mars/snap/kontena-lens/179/.config/Lens"
info: 🐚 Syncing shell environment
info: 💾 Loading stores
STORE MIGRATION (/home/mars/snap/kontena-lens/179/.config/Lens/lens-cluster-store.json): 2.0.0-beta.2
STORE MIGRATION (/home/mars/snap/kontena-lens/179/.config/Lens/lens-cluster-store.json): 2.4.1
STORE MIGRATION (/home/mars/snap/kontena-lens/179/.config/Lens/lens-cluster-store.json): 2.6.0-beta.2
STORE MIGRATION (/home/mars/snap/kontena-lens/179/.config/Lens/lens-cluster-store.json): 2.6.0-beta.3
STORE MIGRATION (/home/mars/snap/kontena-lens/179/.config/Lens/lens-cluster-store.json): 2.7.0-beta.0
STORE MIGRATION (/home/mars/snap/kontena-lens/179/.config/Lens/lens-cluster-store.json): 2.7.0-beta.1
STORE MIGRATION (/home/mars/snap/kontena-lens/179/.config/Lens/lens-cluster-store.json): 3.6.0-beta.1
STORE MIGRATION (/home/mars/snap/kontena-lens/179/.config/Lens/lens-cluster-store.json): 4.2.2

Migrating embedded kubeconfig paths
info: [STORE]: LOADED from /home/mars/snap/kontena-lens/179/.config/Lens/lens-cluster-store.json
info: [STORE]: LOADED from /home/mars/snap/kontena-lens/179/.config/Lens/lens-extensions.json
info: [STORE]: LOADED from /home/mars/snap/kontena-lens/179/.config/Lens/lens-filesystem-provisioner-store.json
STORE MIGRATION (/home/mars/snap/kontena-lens/179/.config/Lens/lens-workspace-store.json): 4.2.0-beta.1
info: [STORE]: LOADED from /home/mars/snap/kontena-lens/179/.config/Lens/lens-workspace-store.json
STORE MIGRATION (/home/mars/snap/kontena-lens/179/.config/Lens/lens-user-store.json): 2.1.0-beta.4
info: [STORE]: LOADED from /home/mars/snap/kontena-lens/179/.config/Lens/lens-user-store.json
info: 🔑 Getting free port for LensProxy server
info: 🔌 Starting LensProxy
info: [LENS-PROXY]: Proxy server has started at http://localhost:45293
info: 🔎 Testing LensProxy connection ...
error: ENOENT: no such file or directory, open '/home/mars/.kube/config' {"errno":-2,"code":"ENOENT","syscall":"open","path":"/home/mars/.kube/config"}
info: ⚡ LensProxy connection OK
info: 🖥️  Starting WindowManager
info: 🧩 Initializing extensions
info: [EXTENSION-DISCOVERY] loading extensions from /home/mars/snap/kontena-lens/179/.config/Lens

(kontena-lens:1373066): libappindicator-WARNING **: 08:48:20.134: Using '/tmp' paths in SNAP environment will lead to unreadable resources
info: [EXTENSION-INSTALLER] installing dependencies at /home/mars/snap/kontena-lens/179/.config/Lens
info: [WINDOW-MANAGER]: Loading Main window from url: http://localhost:45293 ...
info: [EXTENSION-INSTALLER] dependencies installed at /home/mars/snap/kontena-lens/179/.config/Lens
info: [EXTENSION-DISCOVERY] watching extension add/remove in /home/mars/.k8slens/extensions
info: [EXTENSION]: enabled lens-license@0.1.0
info: [STORE]: LOADED from /home/mars/snap/kontena-lens/179/.config/Lens/extension-store/lens-survey/preferences-store.json
info: [EXTENSION]: enabled lens-survey@0.1.0
telemetry main extension activated
info: [STORE]: LOADED from /home/mars/snap/kontena-lens/179/.config/Lens/extension-store/lens-telemetry/preferences-store.json
info: [EXTENSION]: enabled lens-telemetry@0.1.0
info: [WINDOW-MANAGER]: Main window loaded
info: 📡 Checking for app updates
info: Checking for update
error: Error: Error: ENOENT: no such file or directory, open '/snap/kontena-lens/179/resources/app-update.yml'
error: [UPDATE-CHECKER]: failed with an error {"error":"Error: ENOENT: no such file or directory, open '/snap/kontena-lens/179/resources/app-update.yml'"}

sudo apt-get update
sudo apt-get dist-upgrade

$ git clone https://github.com/EdjeElectronics/TensorFlow-Lite-Object-Detection-on-Android-and-Raspberry-Pi.git

ディレクトリー名が長いので、短めな名称にリネーム
$ mv TensorFlow-Lite-Object-Detection-on-Android-and-Raspberry-Pi tflite1
$ cd tflite1

virtualenvを利用した仮想環境を構築：
$sudo pip3 install virtualenv
次のコマンドで仮想環境　"tflite1-env"　を作成
$ python3 -m venv tflite1-env

"tflite1-env"の活性化
$ source tflite1-env/bin/activate

$ bash get_pi_requirements.sh
次のURLから、自分の環境にあったバージョンを選んでインストールする。
https://github.com/google-coral/pycoral/releases/
例えば、python3.8　arm64bitの場合
$pip3 install pip3 install https://github.com/google-coral/pycoral/re
leases/download/v1.0.1/tflite_runtime-2.5.0-cp38-cp38-linux_aarch64.whl

$ wget https://storage.googleapis.com/download.tensorflow.org/models/tflite/coco_ssd_mobilenet_v1_1.0_quant_2018_06_29.zip
$ unzip coco_ssd_mobilenet_v1_1.0_quant_2018_06_29.zip -d Sample_TFLite_model

$ python3 TFLite_detection_video.py --modeldir=Sample_TFLite_model

別の動画を指定する場合,　--videoオプションでファイル名（パス）を指定する。
$ python3 TFLite_detection_video.py --video 動画のファイル名　--modeldir=Sample_TFLite_model

$ python3 TFLite_detection_webcam.py --modeldir=Sample_TFLite_model

import time
import math
import datetime
import cv2
import pigpio
import queue
import numpy as np
import sys
from threading import Thread
import importlib.util
import os

face_cascade_path = '/home/pi/opencv/data/haarcascades/haarcascade_frontalface_default.xml'
face_cascade = cv2.CascadeClassifier(face_cascade_path)
usleep = lambda x: time.sleep(x/1000000.0)

TILT=17
PAN=27
RPi=False
GP=pigpio.pi('localhost',8880)
GP.set_mode(PAN,pigpio.OUTPUT)
GP.set_mode(TILT,pigpio.OUTPUT)

# Define VideoStream class to handle streaming of video from webcam in separate processing thread
# Source - Adrian Rosebrock, PyImageSearch: https://www.pyimagesearch.com/2015/12/28/increasing-raspberry-pi-fps-with-python-and-opencv/
class VideoStream:
    """Camera object that controls video streaming from the Picamera"""
    def __init__(self,resolution=(640,480),framerate=30):
        # Initialize the PiCamera and the camera image stream
        self.stream = cv2.VideoCapture(0)
        #self.stream = cv2.VideoCapture('rtsp://admin:@192.168.68.128:554/1/h264major')
        ret = self.stream.set(cv2.CAP_PROP_FOURCC, cv2.VideoWriter_fourcc(*'MJPG'))
        ret = self.stream.set(3,resolution[0])
        ret = self.stream.set(4,resolution[1])

        # Read first frame from the stream
        (self.grabbed, self.frame) = self.stream.read()

        # Variable to control when the camera is stopped
        self.stopped = False

    def start(self):
        # Start the thread that reads frames from the video stream
        Thread(target=self.update,args=()).start()
        return self

    def update(self):
        # Keep looping indefinitely until the thread is stopped
        while True:
            # If the camera is stopped, stop the thread
            if self.stopped:
                # Close camera resources
                self.stream.release()
                return

            # Otherwise, grab the next frame from the stream
            (self.grabbed, self.frame) = self.stream.read()

    def read(self):
        # Return the most recent frame
        return self.frame

    def stop(self):
        # Indicate that the camera and thread should be stopped
        self.stopped = True
        
def move(p0,p1,dev):
    global tPos,pPos
    global tMin,tMax,pMin,pMax
    if dev==PAN:
        if p1 > pMax or p1 < pMin:
            return
    else:
        if p1 > tMax or p1 < tMin:
            return
            
    deg=p0
    dx=0.4
    counts=int(abs(p1-p0)/dx)
    if p1<p0:
        dx=-dx
    for i in range(0,counts):
        deg=deg+dx
        pw=500+int(deg*2000/270)
        GP.set_servo_pulsewidth(dev,pw)
        #time.sleep(0.005)
        #GP.set_servo_pulsewidth(dev,0)
        if dev==TILT:   
            tPos=deg
        else:
            pPos=deg

def key(k):
    global pPos,tPos,PAN,TILT,track,f_all
    global capture,fontFace,color,Green,Red

    if k == ord('j'):
        new=pPos+2
        move(pPos,new,PAN)
        return
    elif k == ord('k'):
        new=pPos-2
        move(pPos,new,PAN)
        return
    elif k == ord('m'):
        new=tPos-2
        move(tPos,new,TILT)
        return
    elif k == ord('i'):
        new=tPos+2
        move(tPos,new,TILT)
        return
    elif k == ord('p'):
        tmp=input()
        move(pPos,int(tmp),PAN)
    elif k == ord('t'):
        tmp=input()
        move(tPos,int(tmp),TILT)
    elif k == ord('a'):
        f_all = not f_all
    elif k == ord('f'):
        track = not(track)
        if  track:
            color=Red
        else:
            color=Green
 
    elif k == ord('z'):
        move(tPos,0,TILT)
        move(pPos,90,PAN)

def tracking(dX,dY):
    global xW,yW,pPos,tPos,tW
    ret=False
    if dX >0 :
        move(pPos,pPos+1,PAN)
    elif dX < 0:
        move(pPos,pPos-1,PAN)
    if dY > 0:
        move(tPos,tPos+1,TILT)
    elif dY < 0:
        move(tPos,tPos-1,TILT)
    return ret

# 移動体検知
def detectMOV(tm, tc):
    global avg,  img1,frame
    ret = False
    x,y=0,0
    if avg is None:
        avg = img1.copy().astype("float")
    else:
        cv2.accumulateWeighted(img1, avg, 0.5)
        frameDelta = cv2.absdiff(img1, cv2.convertScaleAbs(avg))
        thresh = cv2.threshold(frameDelta, tm,  255, cv2.THRESH_BINARY)[1]
        #cv2.imshow('th',thresh)
        contours,hierarchy = cv2.findContours(thresh, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
    #    contours=cv2.drawContours(img,contours,-1,(0,255,0),2)
        for i in range(0,len(contours)):
            if len(contours[i]) > 0:
                 if cv2.contourArea(contours[i]) > tc:
                    rect = contours[i]
                    x, y, w, h = cv2.boundingRect(rect)
                    cv2.rectangle(frame, (x, y), (x + w, y + h), (0, 0, 255), 2)
                    ret=True
                        
    return ret,x,y

   
def detect_face(frame,gray):
    global xW,yW,xC,yC
    faces = face_cascade.detectMultiScale(gray, scaleFactor=1.3, minNeighbors=5)
    xC,yC=xW,yW
    for x, y, w, h in faces:
        cv2.rectangle(frame, (x, y), (x + w, y + h), (0, 255, 0), 2)
        face = frame[y: y + h, x: x + w]
        try:
            xC,yC=x+w/2,y+h/2
            dX,dY=xC-xW,yC-yW
            if track:
                if(abs(dX)>tW or (abs(dY)>tW)):
                    tracking(dX,dY)
        except:
            xC,yC=xW,yW    
        
if __name__ == "__main__":

    avg=None
    tc=350     # Minimum area  moving detection
    tm=10      # Threshold vale to BINARY
    before = None
    tPos,pPos=0,0
    track=False
    fontFace =cv2.FONT_HERSHEY_SIMPLEX
    Red=(0,0,255)
    Blue=(255,0,0)
    Green=(0,255,0)
    TGT=['airplane','bird','kite']
    #capture = cv2.VideoCapture(1)
    #
    tMin,tMax=0,90  # minimum/Maximum setting for TILT 
    pMin,pMax=0,180 # minimum/Maximum setting for PAN
    move(tPos,0,TILT)
    move(pPos,0,PAN)
    wMax=50
    f_count=wMax

    f_all=True
    MODEL_NAME = 'Sample_TFLite_model'
    GRAPH_NAME = 'detect.tflite'
    LABELMAP_NAME = 'labelmap.txt'
    min_conf_threshold = 0.5
    #resW, resH =1280,720
    resW, resH =640,480
    imW, imH = int(resW), int(resH)
    use_TPU = False
    size=(resW, resH)
    Cx=int(resW/2)
    Cy=int(resH/2)
# Import TensorFlow libraries
# If tflite_runtime is installed, import interpreter from tflite_runtime, else import from regular tensorflow
# If using Coral Edge TPU, import the load_delegate library
    pkg = importlib.util.find_spec('tflite_runtime')
    if pkg:
        from tflite_runtime.interpreter import Interpreter
        if use_TPU:
            from tflite_runtime.interpreter import load_delegate
    else:
        from tensorflow.lite.python.interpreter import Interpreter
        if use_TPU:
            from tensorflow.lite.python.interpreter import load_delegate

    # If using Edge TPU, assign filename for Edge TPU model
    if use_TPU:
        # If user has specified the name of the .tflite file, use that name, otherwise use default 'edgetpu.tflite'
        if (GRAPH_NAME == 'detect.tflite'):
            GRAPH_NAME = 'edgetpu.tflite'

    # Get path to current working directory
    CWD_PATH = os.getcwd()

    # Path to .tflite file, which contains the model that is used for object detection
    PATH_TO_CKPT = os.path.join(CWD_PATH,MODEL_NAME,GRAPH_NAME)

    # Path to label map file
    PATH_TO_LABELS = os.path.join(CWD_PATH,MODEL_NAME,LABELMAP_NAME)

    # Load the label map
    with open(PATH_TO_LABELS, 'r') as f:
        labels = [line.strip() for line in f.readlines()]

    # Have to do a weird fix for label map if using the COCO "starter model" from
    # https://www.tensorflow.org/lite/models/object_detection/overview
    # First label is '???', which has to be removed.
    if labels[0] == '???':
        del(labels[0])

    # Load the Tensorflow Lite model.
    # If using Edge TPU, use special load_delegate argument
    if use_TPU:
        interpreter = Interpreter(model_path=PATH_TO_CKPT,
                                  experimental_delegates=[load_delegate('libedgetpu.so.1.0')])
        print(PATH_TO_CKPT)
    else:
        interpreter = Interpreter(model_path=PATH_TO_CKPT)

    interpreter.allocate_tensors()

    # Get model details
    input_details = interpreter.get_input_details()
    output_details = interpreter.get_output_details()
    height = input_details[0]['shape'][1]
    width = input_details[0]['shape'][2]
    W,H = width,height
    xW,yW =int( W/2),int(H/2)
    tW=W/80         # minimum offcenter distance

    floating_model = (input_details[0]['dtype'] == np.float32)

    input_mean = 127.5
    input_std = 127.5
    move(tPos,20,TILT)
    move(pPos,120,PAN)
    # Initialize frame rate calculation
    frame_rate_calc = 1
    freq = cv2.getTickFrequency()

    # Initialize video stream
    videostream = VideoStream(resolution=(imW,imH),framerate=30).start()
    time.sleep(1)
    frame_rate = 24.0 # フレームレート
    now=datetime.datetime.now().strftime("%Y%m%d_%H%M")
    fmt = cv2.VideoWriter_fourcc('m', 'p', '4', 'v') # ファイル形式(ここではmp4)
    writer = cv2.VideoWriter('SV_'+now+'.mp4', fmt, frame_rate, size) # ライター作成
    frames=0
    #for frame1 in camera.capture_continuous(rawCapture, format="bgr",use_video_port=True):
    while True:
        now=datetime.datetime.now().strftime("%Y%m%d_%H:%M:%S")
        # Start timer (for calculating frame rate)
        t1 = cv2.getTickCount()

        # Grab frame from video stream
        frame1 = videostream.read()

        # Acquire frame and resize to expected shape [1xHxWx3]
        frame = frame1.copy()
        frame_rgb = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
        frame_resized = cv2.resize(frame_rgb, (width, height))
        input_data = np.expand_dims(frame_resized, axis=0)

        # Normalize pixel values if using a floating model (i.e. if model is non-quantized)
        if floating_model:
            input_data = (np.float32(input_data) - input_mean) / input_std

        # Perform the actual detection by running the model with the image as input
        interpreter.set_tensor(input_details[0]['index'],input_data)
        interpreter.invoke()

        # Retrieve detection results
        boxes = interpreter.get_tensor(output_details[0]['index'])[0] # Bounding box coordinates of detected objects
        classes = interpreter.get_tensor(output_details[1]['index'])[0] # Class index of detected objects
        scores = interpreter.get_tensor(output_details[2]['index'])[0] # Confidence of detected objects
        #num = interpreter.get_tensor(output_details[3]['index'])[0]  # Total number of detected objects (inaccurate and not needed)
        # Draw framerate in corner of frame
        msg='FPS: {0:.2f}'.format(frame_rate_calc)
        msg = msg + ' Track:'+str(track)+ ' F:' + str(frames) + ' T:'+ str(f_all) + ' ' + now
        cv2.putText(frame,msg,(30,50),cv2.FONT_HERSHEY_SIMPLEX,0.8,(255,255,0),1,cv2.LINE_AA)
        # Loop over all detections and draw detection box if confidence is above minimum threshold
        for i in range(len(scores)):
            object_name = labels[int(classes[i])] # Look up object name from "labels" array using class index
            if f_all or (object_name in TGT):
                if ((scores[i] > min_conf_threshold) and (scores[i] <= 1.0)):
                    # Get bounding box coordinates and draw box
                    # Interpreter can return coordinates that are outside of image dimensions, need to force them to be within image using max() and min()
                    ymin = int(max(1,(boxes[i][0] * imH)))
                    xmin = int(max(1,(boxes[i][1] * imW)))
                    ymax = int(min(imH,(boxes[i][2] * imH)))
                    xmax = int(min(imW,(boxes[i][3] * imW)))
                    x,y=Cx,Cy
                    if (xmax-xmin)*(ymax-ymin)<10000:
                        x=xmin+int((xmax-xmin)*0.5)
                        y=ymin+int((ymax-ymin)*0.5)
                        if f_all:
                            # Draw label
                            label = '%s: %d%%' % (object_name, int(scores[i]*100))
                            labelSize, baseLine = cv2.getTextSize(label, cv2.FONT_HERSHEY_SIMPLEX, 0.7, 2) # Get font size
                            label_ymin = max(ymin, labelSize[1] + 10) # Make sure not to draw label too close to top of window
                            cv2.rectangle(frame, (xmin,ymin), (xmax,ymax), (10, 255, 0), 2)
                            cv2.rectangle(frame, (xmin, label_ymin-labelSize[1]-10), (xmin+labelSize[0], label_ymin+baseLine-10), (255, 255, 255), cv2.FILLED) # Draw white box to put label text in
                            cv2.putText(frame, label, (xmin, label_ymin-7), cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 0, 0), 2) # Draw label text
                        else:
                            x=xmin+int((xmax-xmin)*0.5)
                            y=ymin+int((ymax-ymin)*0.5)
                            cv2.circle(frame,(x,y),4,color=Green,thickness=1)
                            cv2.circle(frame,(x,y),10,color=Green,thickness=1)
                            cv2.circle(frame,(x,y),16,color=Green,thickness=1)
                            frames=frames+1
                            writer.write(frame)
                            f_count=wMax
                        dW = Cx - x
                        dH = Cy - y
                        msg='dW:'+str(dW)+' dH:'+ str(dH)
                        cv2.putText(frame, msg, (30, 80), cv2.FONT_HERSHEY_SIMPLEX, 0.7, (255, 0, 0), 1) 
                        #msg=tgt_track(track,dH,dW,tH,tW,ptz,moverequest)
                        msg=tracking(dW,dH)

        # All the results have been drawn on the frame, so it's time to display it.
        cv2.imshow('Object detector', frame)
        if f_count>0 and f_count !=wMax:
            if not f_all:
                writer.write(frame)
        f_count=f_count-1
        # Calculate framerate
        t2 = cv2.getTickCount()
        time1 = (t2-t1)/freq
        frame_rate_calc= 1/time1
        # Press 'q' to quit
        k=cv2.waitKey(1) & 0xFF
        key(k)       
        if k == ord('q'):
            break
        for i in range(5):
            frame1 = videostream.read()
    # Clean up
    cv2.destroyAllWindows()
    videostream.stop()
    if writer is not None:
        writer.release()
    move(tPos,10,TILT)
    move(pPos,90,PAN)
    GP.stop()
    print('Finish!')