思维导图备注

Kubeflow 0.6 Document
首页 白天 夜间 BookChat 小程序 小程序 阅读
  • 书签 我的书签
  • 添加书签 添加书签 移除书签 移除书签

Hyperparameter Tuning

 来源:谷歌 浏览 375 扫码 分享 2020-04-19 09:42:07
  • Hyperparameter Tuning
    • Hyperparameter Tuning (Katib)

    Hyperparameter Tuning

    Hyperparameter tuning of ML models in Kubeflow

    Hyperparameter Tuning (Katib)

    Using Katib to tune your model's hyperparameters on Kubernetes

    当前内容版权归 谷歌 或其关联方所有，如需对内容或内容相关联开源项目进行关注与资助，请访问 谷歌 .
    上一篇:
    下一篇:
    • 书签
    • 添加书签 移除书签
    • About
      • Kubeflow
      • Contributing to Kubeflow
      • Community
      • Events Calendar
      • Docs
      • Style Guide for the Kubeflow Docs
    • Getting Started
      • Getting Started with Kubeflow
      • Cloud Installation
        • AWS for Kubeflow
        • Azure for Kubeflow
        • Google Cloud for Kubeflow
        • IBM Cloud Private for Kubeflow
      • Kubernetes Installation
        • Deploying Kubeflow on Existing Clusters
        • Kubeflow Deployment with kfctl_k8s_istio
        • Multi-user, auth-enabled Kubeflow with kfctl_existing_arrikto
      • Workstation Installation
        • Kubeflow on Linux
        • Kubeflow on MacOS
        • Kubeflow on Windows
    • Use Cases
      • GitOps For Kubeflow Using Argo CD
      • Job Scheduling
    • Jupyter Notebooks
      • Overview of Jupyter Notebooks in Kubeflow
      • Set Up Your Notebooks
      • Create a Custom Jupyter Image
      • Submit Kubernetes Resources
      • Build a Docker Image on GCP
      • Troubleshooting Guide
    • Pipelines
      • Pipelines Quickstart
      • Understanding Pipelines
        • Overview of Kubeflow Pipelines
        • Introduction to the Pipelines Interfaces
        • Concepts
        • Pipeline
          • Component
          • Graph
          • Experiment
          • Run and Recurring Run
          • Run Trigger
          • Step
          • Output Artifact
      • Building Pipelines with the SDK
        • Introduction to the Pipelines SDK
        • Install the Kubeflow Pipelines SDK
        • Build Components and Pipelines
        • Manipulate Kubernetes Resources
        • Build Reusable Components
        • Build Lightweight Python Components
        • Best Practices for Designing Components
        • Pipeline Parameters
        • Visualize Results in the Pipelines UI
        • Pipeline Metrics
        • DSL Static Type Checking
        • DSL Recursion
        • GCP-specific Uses of the SDK
        • Enable GPU and TPU
          • Using Preemptible VMs and GPUs on GCP
          • Upgrading and Reinstalling
      • Samples and Tutorials
        • Experiment with the Pipelines Samples
        • Run a Cloud-specific Pipelines Tutorial
      • Reference
        • Component Specification
        • Pipelines API Reference
        • Pipelines SDK Reference
    • Fairing
      • Overview of Kubeflow Fairing
      • Install Kubeflow Fairing
      • Configure Kubeflow Fairing
        • Fairing on GCP
        • Configure Kubeflow Fairing with Access to GCP
      • GCP Samples and Tutorials
      • Train and Deploy on GCP from a Local Notebook
        • Train and Deploy on GCP from a Kubeflow Notebook
          • Tutorials
          • Other Samples and Tutorials
    • Kubeflow on Azure
      • Deployment
      • Install Kubeflow
        • Initial cluster setup for existing cluster
        • Uninstall Kubeflow
        • End-to-End Pipeline Example on Azure
        • Access Control for Azure Deployment
        • Troubleshooting Deployments on Azure AKS
    • Kubeflow on AWS
      • Deployment
      • Install Kubeflow
        • Uninstall Kubeflow
        • Customizing Kubeflow on AWS
        • Logging
        • Private Access
        • Authentication and TLS Support
        • Storage Options
        • Troubleshooting Deployments on Amazon EKS
        • Kubeflow on AWS Features
    • Kubeflow on GCP
      • Deploying Kubeflow
      • Set up a GCP Project
        • Set up OAuth for Cloud IAP
        • Deploy using UI
        • Deploy using CLI
        • Monitor Cloud IAP Setup
        • Delete using CLI
        • Delete using GCP Console
        • Features of Kubeflow on GCP
        • Customizing Kubeflow on GKE
        • Using Your Own Domain
        • Authenticating Kubeflow to GCP
        • Using Cloud Filestore
        • Securing Your Clusters
        • Troubleshooting Deployments on GKE
        • End-to-end Kubeflow on GCP
        • Pipelines End-to-end on GCP
        • Logging and monitoring
    • Components of Kubeflow
      • Jupyter Notebooks
      • Hyperparameter Tuning
      • Hyperparameter Tuning (Katib)
      • Pipelines
      • Pipelines
      • Serving
      • Istio Integration (for TF Serving)
        • Seldon Serving
        • NVIDIA TensorRT Inference Server
        • TensorFlow Serving
        • TensorFlow Batch Predict
        • PyTorch Serving
      • Training
      • Chainer Training
        • MPI Training
        • MXNet Training
        • PyTorch Training
        • TensorFlow Training (TFJob)
      • Miscellaneous
      • Metadata
        • Nuclio functions
    • Tutorials, Samples, and Shared Resources
      • Kubeflow Samples
      • Codelabs, Workshops, and Tutorials
      • Blog Posts
      • Videos
      • Shared Resources and Components
    • Further Setup and Troubleshooting
      • Accessing Kubeflow UIs
      • Virtual Developer Environments
      • Microk8s for Kubeflow
        • MiniKF
        • Minikube for Kubeflow
        • Configuring Kubeflow with kfctl and kustomize
        • Kubeflow On-prem in a Multi-node Kubernetes Cluster
        • Usage Reporting
        • Multi-user Isolation
        • Troubleshooting
    • Upgrading Kubeflow
      • Upgrading a Kubeflow Deployment
    • Reference
      • Reference Overview
      • PyTorchJob Reference
        • PyTorchJob v1
        • PyTorchJob
        • PyTorchJob v1beta2
        • PyTorchJob
      • TFJob Reference
        • TFJob v1
        • TFJob Common
          • TFJob TensorFlow
        • TFJob v1beta2
        • TFJob Common
          • TFJob TensorFlow
      • Dockerfile Locations
      • Metadata API Reference
    暂无相关搜索结果！

      本文档使用 BookStack 构建

      展开/收起文章目录

      分享，让知识传承更久远

      文章二维码

      手机扫一扫，轻松掌上读

      文档下载

      • 普通下载
      • 下载码下载(免登录无限下载)
      你与大神的距离，只差一个APP
      APP下载
      请下载您需要的格式的文档，随时随地，享受汲取知识的乐趣！
      PDF文档 EPUB文档 MOBI文档
      温馨提示 每天每在网站阅读学习一分钟时长可下载一本电子书，每天连续签到可增加阅读时长
      下载码方式下载：免费、免登录、无限制。 免费获取下载码

      微信小程序阅读

      BookChat 微信小程序阅读
      您与他人的薪资差距，只差一个随时随地学习的小程序

      书签列表

        阅读记录

        阅读进度: 0.00% ( 0/0 ) 重置阅读进度