什么是 YAML 文件，它在机器学习环境中的用途是什么？

数据挖掘 机器学习 神经网络 深度学习 序列化
2021-10-05 02:59:09

我不完全确定这是否是主题，所以如果不是，请告诉我。我在阅读机器学习文献时不断看到 YAML 文件的想法。我的问题是，YAML 文件到底是什么，它与机器学习和数据科学项目有什么关系？

2个回答

有很多方法可以序列化数据。例如，JSON、XML 和 CSV 是可能的序列化方法。另一种方式是 YAML。它是人类可读的，这意味着它包含使用常用单词编写的可读文本。这反对二进制序列化。每当您打开文件时，您将无法看到写入的内容，因为它是以二进制格式序列化的，而不是人类可读的格式。您可能会在这里找到它，它主要用于配置文件。在 ML 上下文中，根据您的任务，您可能会面临不同的文件格式。例如，有监督的 ML 任务广泛使用的格式之一是 CSV，尽管您可能会发现其他人使用他们最喜欢的编码。您可能还想进行自己的序列化，这在公司内部操作中很少见。

除了数据之外，yaml 文件通常用作人类可读的配置文件……这里是一个 docker compose yaml 文件的示例

# version: "3.5"
version: '3'
# version: '2'
services:


  nodejs-admin:
    image: ${GKE_APP_IMAGE_ADMIN}
	    hostname: admin
	    container_name: loud_admin
	    restart: always
	    depends_on:
	      - loudmongo
	      # - loudmail
	      - loudmail
	    volumes:
	      - /cryptdata4/var/log/loudlog-admin:/loudlog-admin
	      - /cryptdata5/var/log/blobs:/blobs
	      - /cryptdata5/var:/cryptdata5/var
	      - /cryptdata5/var/tools:/tools
	      - /cryptdata6/var/log/loudlog-enduser:/loudlog-enduser
	      # - ${TMPDIR_GRAND_PARENT}/curr/loud-build/${PROJECT_ID}/webapp/admin/bundle:/tmp
	      - ${TMPDIR_GRAND_PARENT}/${bundleNormal}/loud-build/${PROJECT_ID}/webapp/admin/bundle:/tmp
      # - $SOURCE_REPO_DIR/tests:/tmp/tests
	    environment:
	      - MONGO_SERVICE_HOST=loudmongo
	      - MONGO_SERVICE_PORT=$GKE_MONGO_PORT
      - MONGO_URL=mongodb://loudmongo:$GKE_MONGO_PORT/admin
	      - METEOR_SETTINGS=${METEOR_SETTINGS}
      # when sending to port 587 will NOT get routed through email content filter however port 25 will get routed
      - MAIL_URL=smtp://support@${GKE_DOMAIN_NAME}:ignore_this@loudmail:587/
	      # - MAIL_URL=smtp://support@${GKE_DOMAIN_NAME}:ignore_this@loudmail:25/
      - GKE_NOTIF_TASK_OVERDUE=$GKE_NOTIF_TASK_OVERDUE
	      - GKE_NOTIF_PLANS_RECUR=$GKE_NOTIF_PLANS_RECUR
      - GKE_NOTIF_SHIFTS_RUN=$GKE_NOTIF_SHIFTS_RUN
	      - GKE_NOTIF_LOAD_RUN=$GKE_NOTIF_LOAD_RUN
      - GKE_NOTIF_ORGS_HEALTH_RUN=$GKE_NOTIF_ORGS_HEALTH_RUN
	    links:
	      - loudmongo
	      # - loudmail
	      - loudmail
	    ports:
	     - 127.0.0.1:${PORT_ADMIN}:3001
#   networks:
#    - loudthink-network
    working_dir: /tmp
    command: /usr/bin/supervisord -c /etc/supervisor/conf.d/supervisord.conf

  nodejs-enduser:
    # image: ${GKE_APP_REPO_PREFIX}/${PROJECT_ID}/loudweb-enduser
    image: ${GKE_APP_IMAGE_ENDUSER}
	    hostname: enduser
	    container_name: loud_enduser
	    restart: always
	    depends_on:
	      - nodejs-admin
	      - loudmongo
	      # - loudmail
	      - loudmail
	    volumes:
	      - /cryptdata6/var/log/loudlog-enduser:/loudlog-enduser
	      - /cryptdata5/var/log/blobs:/blobs
	      # - ${TMPDIR_GRAND_PARENT}/curr/loud-build/${PROJECT_ID}/webapp/enduser/bundle:/tmp
	      # - ${TMPDIR_GRAND_PARENT}/curr/loud-build/${PROJECT_ID}/webapp/admin/bundle/programs/server/assets/app/config/apn-cert.pem:/private/config/apn-cert.pem
	      # - ${TMPDIR_GRAND_PARENT}/curr/loud-build/${PROJECT_ID}/webapp/admin/bundle/programs/server/assets/app/config/apn-key.pem:/private/config/apn-key.pem
	      - ${TMPDIR_GRAND_PARENT}/${bundleNormal}/loud-build/${PROJECT_ID}/webapp/enduser/bundle:/tmp
      - ${TMPDIR_GRAND_PARENT}/${bundleNormal}/loud-build/${PROJECT_ID}/webapp/admin/bundle/programs/server/assets/app/config/apn-cert.pem:/private/config/apn-cert.pem
	      - ${TMPDIR_GRAND_PARENT}/${bundleNormal}/loud-build/${PROJECT_ID}/webapp/admin/bundle/programs/server/assets/app/config/apn-key.pem:/private/config/apn-key.pem
    environment:
      - MONGO_SERVICE_HOST=loudmongo
      - MONGO_SERVICE_PORT=$GKE_MONGO_PORT
	      - MONGO_URL=mongodb://loudmongo:$GKE_MONGO_PORT/admin
      - METEOR_SETTINGS=${METEOR_SETTINGS}
	      # when sending to port 587 will NOT get routed through email content filter however port 25 will get routed
	      - MAIL_URL=smtp://support@${GKE_DOMAIN_NAME}:ignore@loudmail:587/
      # - MAIL_URL=smtp://support@${GKE_DOMAIN_NAME}:ignore@loudmail:25/
	    links:
	      - loudmongo
	      # - loudmail
	      - loudmail
	    ports:
	     - 127.0.0.1:${PORT_ENDUSER}:3000
#   networks:
#    - loudthink-network
    working_dir: /tmp
    command: /usr/bin/supervisord -c /etc/supervisor/conf.d/supervisord.conf

  loudmongo:
    # image: mongo
    # image: mongo:3.6
    image: $GKE_MONGO_IMAGE
	    hostname: mongo
	    container_name: loud_mongo
	    restart: always
	    ports:
	     - 127.0.0.1:$GKE_MONGO_PORT:$GKE_MONGO_PORT
#   logpath:
#    - /data/logs/mongo.log
    volumes:
     - /cryptdata7/var/data/db:/data/db
     # - /cryptdata7/var/data/logs:/data/logs




  loud-devops:
    # image: dind
    image: localhost:5000/hygge/loudweb-dind/1904082202
    hostname: devops
    container_name: loud_devops
    restart: always
    ports:
      - 127.0.0.1:9000:9000
#     expose:
#      - 9000
#   networks:
#    - loudthink-network
#     environment:
#       - SSH_AUTH_SOCK=$SSH_AUTH_SOCK
	    volumes:
	      - /var/run/docker.sock:/var/run/docker.sock   
	      - /usr/bin/docker:/usr/bin/docker   
	      # - $SSH_AUTH_SOCK:/ssh-agent 
      - /usr/lib/x86_64-linux-gnu/libltdl.so.7:/usr/lib/x86_64-linux-gnu/libltdl.so.7  
      - /usr/lib/x86_64-linux-gnu/libgpm.so.2:/usr/lib/x86_64-linux-gnu/libgpm.so.2
      - /home/khepri/src/github.com/loudthink:/inner_home/khepri/src/github.com/loudthink 
      - /home/khepri/.docker:/inner_home/khepri/.docker
      - /cryptdata5/var/tools/inner_home/khepri:/inner_home/khepri
      - $GKE_DIND_SUPERVISOR_LOG_DIR:/var/log/supervisor
	      - /cryptdata5/var/tools/usr/local/go:/usr/local/go
	      - /cryptdata6/var/log/tmp/khepri01:/cryptdata6/var/log/tmp/khepri01
	      - /cryptdata6/var/log/tmp/shared:/cryptdata6/var/log/tmp/shared
	      - /cryptdata5/var/tools/usr/local/bin:/usr/local/bin 
	      - /cryptdata:/cryptdata
	      - /cryptdata2:/cryptdata2
	      - /cryptdata4:/cryptdata4
	      - /cryptdata5:/cryptdata5
	      - /cryptdata6:/cryptdata6
	      - /cryptdata7:/cryptdata7
	      # - /etc/letsencrypt/live:/etc/letsencrypt/live
	      # - /etc/letsencrypt/archive/medssenger-dev.medstack.net:/etc/letsencrypt/archive/medssenger-dev.medstack.net
	      - /usr/local/ssl:/usr/local/ssl
	      #  following line HOME is only used during install as referenced in loudspeed/build/bin/setup_devops_sudo_cmds.sh to copy over .ssh files
	      # - ${HOME}:${HOME}
      # - /var/lib/docker:/var/lib/docker
    command: /usr/bin/supervisord -c /etc/supervisor/supervisord.conf



  loudmail:
    image: tvial/docker-mailserver:latest
    hostname: mail
    domainname: ${GKE_DOMAIN_NAME}
	    container_name: loud_mail
	    restart: always
	    environment:
	      - PERMIT_DOCKER=network
	      - SSL_TYPE=letsencrypt
	      - ONE_DIR=1
	      - DMS_DEBUG=1
	      - SPOOF_PROTECTION=0
	      - REPORT_RECIPIENT=1
	      - ENABLE_SPAMASSASSIN=0
	      - ENABLE_CLAMAV=0
	      - ENABLE_FAIL2BAN=1
	      - ENABLE_POSTGREY=0
	    cap_add:
	      - NET_ADMIN
	      - SYS_PTRACE
	    ports:
	    - "25:25"
	    - "587:587"
	    - "465:465"
	    volumes:
	      - ${GKE_MAIL_EHOOK_DIR}/data/:/var/mail/
      # - ${GKE_MAIL_EHOOK_DIR}/state/:/var/mail-state/
	      - ${GKE_MAIL_EHOOK_DIR}/state/:${GKE_MAIL_EHOOK_INNER_STATE_DIR}/
	      - ${GKE_MAIL_EHOOK_DIR}/config/:/tmp/docker-mailserver/
      - ${GKE_MAIL_EHOOK_OUTPUT_DIR}/:${GKE_MAIL_EHOOK_INNER_OUTPUT_DIR}/
      - ${GKE_MAIL_EHOOK_HOST_DIR}/:/local_ehook/
	      # - /etc/letsencrypt/:/etc/letsencrypt/
	      - ${GKE_LETSENCRYPT_DIR}/:/etc/letsencrypt/
      - ${GKE_MAIL_EHOOK_DIR}/log/:/var/log/mail/

与任何配置文本文件一样，模板的概念很有用......例如，如果您的名为 GKE_COMPOSE_YAML 的 yaml 文件包含 say 

foobar: some_token

some_token您可以通过在预处理阶段进行字符串搜索和替换来更新硬编码

sed -i -- "s|some_token|${my_current_token}|g"   $GKE_COMPOSE_YAML

根据您的进程用于解析 yaml 文件的内容，它可能允许按照 yaml 文件中的以下代码段将环境变量嵌入到您的 yaml 文件中 

    foobar: $my_current_token

当您的环境包含可能已使用定义的 var my_current_token 时

export my_current_token=something.cool.here
其它你可能感兴趣的问题
上一篇如何将用于训练的自定义数字特征添加到 BERT 微调？ 下一篇术语：聚类上下文中的“平面几何”