README_Template.md

MODEL

The following is a brief directory structure and description for this example:

├── data                          # Data set directory
│   └── README.md                   # Documentation describing how to prepare dataset
├── distribute_k8s                # Distributed training related files
│   ├── distribute_k8s_BF16.yaml    # k8s yaml to crate a training job with BF16 feature
│   ├── distribute_k8s_FP32.yaml    # k8s yaml to crate a training job
│   └── launch.py                   # Script to set env for distributed training
├── README.md                     # Documentation
├── result                        # Output directory
│   └── README.md                   # Documentation describing output directory
└── train.py                      # Training script

Content

• MODEL
  • Content
  • Model Structure
  • Usage
    • Stand-alone Training
    • Distribute Training
  • Benchmark
    • Stand-alone Training
      • Test Environment
      • Performance Result
    • Distribute Training
      • Test Environment
      • Performance Result
  • Dataset
    • Prepare
    • Fields
    • Processing
  • TODO LIST

Model Structure

What is this model
The discription of model in this folder(such as params, model size)
The structure of the model

Usage

How to use MODEL model example The first step in following chapter should be data preparation
Put a link to #dataset, Like Data Prepare

Stand-alone Training

How to train stand-alone model

Distribute Training

How to train distribute model

Benchmark

The benchmark of example

Stand-alone Training

The benchmark of stand-alone training

Test Environment

The information of hardware & software stand-alone test environment

Performance Result

	Framework	DType	Accuracy	AUC	Globalsetp/Sec
Model Type	Community TensorFlow	FP32			(baseline)
	DeepRec w/ oneDNN	FP32			(+1.00x)
	DeepRec w/ oneDNN	FP32+BF16			(+1.00x)

• Community TensorFlow version is v1.15.5.

Distribute Training

The benchmark of distribute training

Test Environment

The information of hardware & software stand-alone test environment

Performance Result

	Framework	Protocol	DType	Globalsetp/Sec
Model Type	Community TensorFlow	GRPC	FP32
	DeepRec w/ oneDNN	GRPC	FP32
	DeepRec w/ oneDNN	GRPC	FP32+BF16

• Community TensorFlow version is v1.15.5.

Dataset

Whic dataset is used

Prepare

Where to download dataset
Where to put dataset
Here put a link to data/README.md

Fields

A detailed description of dataset

Processing

How data are processed in this example

TODO LIST

Next To do

• Pending to do sth