TensorFlow Model Garden Documentation (Beta/Experimental)

The TensorFlow Model Garden (TFMG) is a collection of models that use TFMG’s high-level APIs. They are intended to be well-maintained, tested, and kept up to date with the latest TFMG API.

The goal of the TFMG is to develop exemplary implementations of prominent machine learning models in community. They should also be reasonably optimized for fast performance while still being easy to read.

These models are used as end-to-end tests, ensuring that the models run with the same or improved speed and performance with each new TensorFlow build.

This documentation is to explain a process for reproducing a state-of-the-art machine learning model at a level of quality suitable for inclusion in the TFMG, which includes the engineering process and elaborate on each step, from paper analysis to model release.

Overview

• Modeling Libraries
  • official.modeling.tf_utils
  • official.modeling.performance
  • official.modeling.grad_utils
  • TF-NLP for Natural Language Processing
    • modeling.tf_utils
  • TF-Vision for Computer Vision
• Training Library
  • Code WorkFlow
    • Task
    • Training Binary
    • Experiment Configuration
    • Experiment Registry
    • Overriding Configuration via Yaml and Gin
  • Run locally

Quick Start

• Prerequisites
• Installation
  • Requirements
  • Installation

Model Zoo

• TF-Vision Model Garden
  • Introduction
  • Image Classification
  • Object Detection and Instance Segmentation
  • Semantic Segmentation
  • Video Classification
• TF-NLP Model Zoo

Tutorials

• Learn about Configs
  • flags command line args
• Customize Dataset
  • Datasets
• Customize Data Pipelines
  • Extract-Transform-Load (ETL) Data Pipeline
• Customize Models
  • TFMG Customize Model Structure
  • official.modeling.tf_utils
• Customize Losses
• Finetuning Models

Colab Example: TF-Vision

• Guides for Creating TF-Vision Project
  • Create Model
  • Create Dataloader
  • Create Config
  • Create Task
  • Import registry
  • Training
• TF-Vision Domain-Specific Example

Colab Example: TF-NLP

• NLP Modeling Library
• TF-NLP Domain-Specific Example

Contribution

• Contributions