Welcome to the TensorFlow Extended (TFX) tutorials page! TFX is an end-to-end platform for automating machine learning (ML) workflows. This page will guide you through various TFX tutorials to help you get started with TFX and its capabilities.

快速开始

If you are new to TFX, we recommend starting with the following tutorials:

教程列表

Here is a list of tutorials available for TFX:

TFX 概述

TFX is a comprehensive platform that provides tools for building, training, and deploying ML models at scale. It helps organizations streamline their ML workflows and ensures consistency and reproducibility.

  • 核心组件: TFX includes components like TFX Orchestration, TFX Triggers, and TFX Datasets.
  • 优势: TFX helps in automating repetitive tasks, reducing manual errors, and improving collaboration among ML teams.

TFX 工作流程

Understanding the TFX workflow is crucial for implementing TFX in your ML projects. The workflow typically involves the following steps:

  1. Data Ingestion: Collect and prepare data for ML training.
  2. Feature Engineering: Transform and preprocess data to create features.
  3. Model Training: Train models using the prepared data.
  4. Model Evaluation: Evaluate the trained models for performance.
  5. Model Deployment: Deploy the best models into production.

TFX 与 KFP 的集成

TensorFlow Extended (TFX) 与 Kubernetes Flow (KFP) 是两个强大的 ML 工具。它们可以无缝集成，以提供更强大的 ML 工作流程自动化。

  • 集成优势: 集成 TFX 和 KFP 可以实现更灵活、可扩展的 ML 工作流程。
  • 教程链接: 如何集成 TFX 和 KFP

TFX 在生产环境中的应用

TFX 在生产环境中具有广泛的应用。以下是一些关键场景：

  • 自动化模型部署: 使用 TFX 自动化模型的部署过程，确保模型的快速上线。
  • 监控和日志记录: 利用 TFX 提供的监控和日志记录功能，跟踪模型的性能和状态。

TensorFlow Extended Architecture

For more detailed information and tutorials, please visit our TFX 官方文档.

希望这些教程能帮助您更好地了解和使用 TFX。如果您有任何问题或建议，请通过 联系我们 页面与我们联系。