TAO Toolkit Quick Start Guide

This page provides a video and text-based quick start guide for installing and running TAO Toolkit.

Hardware Requirements

Minimum System Configuration

• 8 GB system RAM

• 4 GB of GPU RAM

• 8 core CPU

• 1 NVIDIA GPU

• 100 GB of SSD space

Recommended System Configuration

• 32 GB system RAM

• 32 GB of GPU RAM

• 8 core CPU

• 1 NVIDIA GPU

• 100 GB of SSD space

TAO Toolkit is supported on discrete GPUs, such as H100, A100, A40, A30, A2, A16, A100x, A30x, V100, T4, Titan-RTX, and Quadro-RTX.

Software Requirements

Getting Started

Download the TAO package that contains startup scripts, Jupyter Notebooks, and config files. TAO is supported on Google Colab; if you want to try on Colab, you can skip this step and scroll down to the running in colab section.

            

            
wget --content-disposition https://api.ngc.nvidia.com/v2/resources/nvidia/tao/tao-getting-started/versions/5.3.0/zip -O getting_started_v5.3.0.zip
unzip -u getting_started_v5.3.0.zip  -d ./getting_started_v5.3.0 && rm -rf getting_started_v5.3.0.zip && cd ./getting_started_v5.3.0
        

File Hierarchy and Overview

The TAO Toolkit getting started resource is broadly classified into two components.

• setup: These are a set of quick start scripts to help install and deploy the TAO Toolkit launcher and the TAO Toolkit API’s in various Cloud Service Providers.

• notebooks: These are beginner friendly end-to-end tutorial notebooks, that help you hit the ground running with TAO. The notebooks install TAO, download the required data, and run TAO commands end-to-end, for various use cases.

  These notebooks are broadly classified into the following 3 categories.

  • tao_api_starter_kit: These are end-to-end notebooks that help you learn the features

    supported by the TAO Toolkit API model of execution. The notebooks under the api directory, work directly at the REST API level using REST API requests, while the client directory uses the TAO Client CLI to interact with the API server.

  • tao_launcher_starter_kit: This directory includes sample notebooks that walk you through the end-to-end workflow for all the computer vision models

    supported in TAO Toolkit. You can interact with the Toolkit using the TAO Toolkit launcher CLI.

  • tao_data_services: This directory includes sample notebooks that walk you through the end-to-end workflow of the different

    dataset manipulation and annotation tools that are included as part of the TAO Toolkit.

            

            
setup

   |--> quickstart_launcher.sh
   |--> quickstart_api_bare_metal
   |--> quickstart_api_aws_eks
   |--> quickstart_api_azure_aks
   |--> quickstart_api_gcp_gke

notebooks

   |--> tao_api_starter_kit

      |--> api

            |--> automl
            |--> end2end
            |--> dataset_prepare

      |--> client

            |--> automl
            |--> end2end
            |--> dataset_prepare

   |--> tao_launcher_starter_kit

      |--> dino
      |--> deformable_detr
      |--> classification_pyt
      |--> ocdnet
      |-->  ...

   |--> tao_data_services

      |--> data
      |-->  ...
        

TAO toolkit is available as a Docker container or a collection of Python wheels. You can run the TAO Toolkit in one of the following ways:

• the launcher CLI

• the containers directly

• the TAO Toolkit APIs

• Python wheels

Launcher CLI

The TAO Toolkit Launcher is a lightweight Python based command-line interface. The launcher acts as a front-end for TAO Toolkit containers built on top of PyTorch and TensorFlow. The CLI abstracts you from the information about which network is implemented in what container. The respective container gets launched automatically based on the model you plan to use.

To get started with the launcher, use the follow the instructions to install the required pre-requisite software.

Installing the Pre-requisites

The TAO Toolkit Launcher is strictly a Python3 only package, capable of running on Python versions >= 3.7.

1. Install docker-ce by following the official instructions.

   After you have installed docker-ce, follow the post-installation steps to ensure that the Docker can be run without sudo.

2. Install nvidia-container-toolkit by following the install-guide.

3. Get an NGC account and API key:

   1. Go to NGC and click the TAO Toolkit container in the Catalog tab. This message is displayed: “Sign in to access the PULL feature of this repository”.

   2. Enter your Email address and click Next, or click Create an Account.

   3. Choose your organization when prompted for Organization/Team.

   4. Click Sign In.

4. Log in to the NGC Docker registry (nvcr.io) using the command docker login nvcr.io and enter the following credentials:

   Copy

   Copied!

               
   
               
   a. Username: "$oauthtoken"
   b. Password: "YOUR_NGC_API_KEY"
           

   
   

   where YOUR_NGC_API_KEY corresponds to the key you generated from step 3.

5. Setup your Python environment with Python version >= 3.7

   Note

   NVIDIA recommends setting up a Python environment using miniconda. The following instructions show how to setup a Python conda environment.

   1. Follow the instructions in this link to set up a Conda environment using Miniconda.

   2. After you have installed miniconda, create a new environment by setting the Python version to 3.7.

      Copy

      Copied!

                  
      
                  
      conda create -n launcher python=3.7
              

   3. Activate the conda environment that you have just created.

      Copy

      Copied!

                  
      
                  
      conda activate launcher
              

   4. Once you have activated your conda environment, verify that the command prompt shows the name of your Conda environment.

      Copy

      Copied!

                  
      
                  
      (launcher) py-3.7 desktop:
              

   5. When you are done with your session, you can deactivate your conda environment using the deactivate command:

      Copy

      Copied!

                  
      
                  
      conda deactivate
              

   6. You may re-instantiate this conda environment using the following command.

   Copy

   Copied!

               
   
               
   conda activate launcher
           

   
   

Installing TAO Launcher

After you have installed the required pre-requisites:

1. Install the CLI launcher using the quick start script downloaded with the getting_started NGC package from here

   Copy

   Copied!

               
   
               
   bash setup/quickstart_launcher.sh --install
           

2. Use this script to update the launcher to the latest version of TAO Toolkit by running the following command:

   Copy

   Copied!

               
   
               
   bash setup/quickstart_launcher.sh --upgrade
           

3. Invoke the entry points using the tao command.

   Copy

   Copied!

               
   
               
   tao --help
           

   The sample output of the above command is:

   Copy

   Copied!

               
   
               
   usage: tao [-h] {list,stop,info,dataset,deploy,model} ...
   
   Launcher for TAO
   
   optional arguments:
   -h, --help            show this help message and exit
   
   task_groups:
         {list,stop,info,dataset,deploy,model}
           

   Under tasks you can see all the launcher-invokable tasks. The following are the specific tasks that help with handling the launched commands using the TAO Launcher:

   • list

   • stop

   • info

            

            
export PATH=$PATH:~/.local/bin
        

Running from Container

You can run TAO directly using the Docker container. To use the container directly, you need to know which container to pull. There are multiple containers under TAO, and depending on the model that you want to train you must pull the appropriate container. This is not required when using the Launcher CLI.

            

            
export DOCKER_REGISTRY="nvcr.io"
export DOCKER_NAME="nvidia/tao/tao-toolkit"
export DOCKER_TAG="5.0.0-tf1.15.5" ## for TensorFlow docker
export DOCKER_CONTAINER=$DOCKER_REGISTRY/$DOCKER_NAME:$DOCKER_TAG

docker run -it --rm --gpus all -v /path/in/host:/path/in/docker $DOCKER_CONTAINER \
       detectnet_v2 train -e /path/to/experiment/spec.txt -r /path/to/results/dir -k $KEY --gpus 4
        

For detailed instructions on how to run directly from containers, refer to this section.

Running TAO Toolkit APIs

TAO Toolkit API is a Kubernetes service that enables building end-to-end AI models using REST APIs. The API service can be installed on a Kubernetes cluster (local / AWS EKS) using a Helm chart along with minimal dependencies. TAO Toolkit jobs can be run using the GPUs available on the cluster and can scale to a multi-node setting. You can use a TAO client CLI to interact with TAO services remotely or you can integrate it with your apps and services directly using REST APIs.

To get started, use the provided one-click deploy script to deploy either on bare-metal setup or on managed Kubernetes service like Amazon EKS. Jupyter Notebooks to train using the APIs directly or using the client app is provided under notebooks/api_starter_kit.

            

            
bash setup/quickstart_api_bare_metal/setup.sh install
bash setup/quickstart_api_aws_eks/setup.sh install
        

More information about setting up the API services and the API is provided

here.

Running from Python Wheels

You can also run TAO directly on bare-metal without Docker or K8s by using the Python wheels. These Python wheels contain stand alone implementations of the DNN functionality that are pre-built and packaged into the TAO Toolkit containers.

The table below helps map the TAO Toolkit wheel to the container and captures any exceptions about these wheels.

TAO Toolkit provides sample tutorials to deploy directly on Google Colab without having to configure your infrastructure. The full instructions on how to work with Google Colab are provided here.

Installing nvidia_tao_deploy locally

The section below details the instructions on how to install the nvidia_tao_deploy wheels locally

1. Install the following Python pip dependencies:

   Copy

   Copied!

               
   
               
   python3 -m pip install --upgrade pip
   python3 -m pip install Cython==0.29.36
   python3 -m pip install nvidia-ml-py
   python3 -m pip install nvidia-pyindex
   python3 -m pip install --upgrade setuptools
   python3 -m pip install pycuda==2020.1
   python3 -m pip install nvidia-eff-tao-encryption
   python3 -m pip install nvidia-eff
   python3 -m pip install cffi
           

2. Setup openMPI and mpi4py:

   Copy

   Copied!

               
   
               
   sudo apt-get install libopenmpi-dev -y
   python3 -m pip install mpi4py
           

3. Install the TAO wheel:

   Copy

   Copied!

               
   
               
   python3 -m pip install nvidia-tao-deploy
           

Installing nvidia_tao_pytorch Locally

The nvidia-tao-pytorch wheel has several 3rd party dependencies, which can be quite cumbersome to install. Inorder to help with the build installation, please follow the steps in this script:

As a part of the TAO Toolkit package, example Jupyter Notebooks are available for all the tasks that are supported in TAO. The Notebooks broadly cover the following steps:

1. Setting up the local enviroment for TAO.

2. Installing the TAO Toolkit launcher CLI.

3. Download the dataset and pre-trained model.

4. Running the end-to-end workflow of the computer vision task. A brief description of the Notebook is covered in the tables below.

Computer Vision

All computer vision samples are included in the getting started resource on NGC.

Purpose-Built Pre-Trained Models

The following is a list of purpose-built pre-trained models mapped with their corresponding samples.

Open model architecture

The TAO Toolkit Docker gives you access to a repository of pretrained models that can serve as a starting point when training deep neural networks. These models are hosted on the NGC. To download the models:

1. Download the NGC CLI and install it. More information about the NGC Catalog CLI is available here.

2. Follow the instructions below to configure the NGC CLI and download the models.

Listing all Available Models

Use this command to get a list of models that are hosted in the NGC model registry:

            

            
ngc registry model list <model_glob_string>
        

Here is an example of using this command for the computer vision models:

            

            
ngc registry model list nvidia/tao/pretrained_*
        

This command gives us a list of the pretrained backbones available for different tasks:

• Classification

• Object Detection with Detectnet_v2

• Object Detection with SSD/DSSD/YOLOv3/YOLOv4/YOLOv4-Tiny/FasterRCNN/RetinaNet

• Object Detection with EfficientDet

• Instance Segmentation

• Semantic Segmentation

To view the full list of computer vision models, use the following command:

            

            
ngc registry model list nvidia/tao/*
        

Downloading a Model

Use this command to download the model you have chosen from the NGC model registry:

            

            
ngc registry model download-version <org/team/model_name:version> -dest <path_to_download_dir>
        

For example, use this command to download the resnet 18 classification model to the $USER_EXPERIMENT_DIR directory:

            

            
ngc registry model download-version
nvidia/tao/pretrained_classification:resnet18 --dest $USER_EXPERIMENT_DIR/pretrained_resnet18
        

            

            
Downloaded 82.41 MB in 9s, Download speed: 9.14 MB/s
----------------------------------------------------
Transfer id: pretrained_classification_vresnet18 Download status: Completed.
Downloaded local path: /workspace/tao-experiments/pretrained_resnet18/
Total files downloaded: 2
Total downloaded size: 82.41 MB
Started at: 2019-07-16 01:29:53.028400
Completed at: 2019-07-16 01:30:02.053016
Duration taken: 9s seconds
        

Once you have downloaded the notebook samples and required pretrained models, you can start the respective sample notebook with the following command:

            

            
jupyter notebook --ip 0.0.0.0 --port 8888 --allow-root
        

Open an internet browser on localhost and navigate to the following URL:

            

            
http://0.0.0.0:8888
        

Execute the cells in the notebook to train a model using TAO Toolkit.

After training is complete, follow these instructions to deploy a computer vision model to DeepStream.