This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TDA4VM: Jacinto TI Edge AI Monthly Webinar (Jul 2021): Embedded deep learning deployment. Demystified.

Part Number: TDA4VM

In this webinar, the third in our monthly series on TI Edge AI technology, you will learn how TI Edge AI processors and software solutions let you maximize AI model performance on specialized deep learning accelerators without having to program the cores. Our solutions let you deploy AI models to your embedded applications with industry-standard APIs (e.g. TensorFlow Lite, ONNX Runtime and TVM). This approach combines the performance advantages of TI Edge AI processors with a programming environment that is simple, flexible, and easy to use.

Webinar topics:

  • Visualizing deep learning models to understand how they operate on an embedded processor
  • Optimization techniques to maximize performance
  • Using industry-standard APIs to compile, deploy, and accelerate your models
  • Hands-on session to explore various deployment techniques
  • TI Model Zoo—60+ pre-compiled models to help you develop faster and more efficiently
  • Performance benchmarking methodologies—what you need to know


Register below for one of the two sessions depending on your timezone.

Session1: Wednesday, July 14, 2021 10:00 am, Central Daylight Time (Chicago, GMT-05:00)
Session2: Wednesday, July 14, 2021 9:00 pm, Central Daylight Time (Chicago, GMT-05:00)


it is recommended to watch the previous webinar to understand the development flow at the link below.

training.ti.com/build-edge-ai-hello-world-application-using-free-online-tools

Please feel free to post any questions below for preparation or if you have any difficulty running the previous webinar code.

  • Thank you all for your interest in the webinar.

    The attached ZIP file has code examples and other required scripts for you to reproduce the results either during the webinar or after at your convenience.

    We will also post the webinar video recording after the event is over.

    Directory Contents
    ==================
    - README: this file
    - other files and scripts needed for the program.

    Model compilation and deployment options
    =================================

    You can run the below directly on the EdgeAI cloud tool: https://dev.ti.com/edgeai/

    Example #1: 3_EdgeAI_Model_Compile_Run_accu_1_20_calib.ipynb

    Example #2: 3_EdgeAI_Model_Compile_Run_16b_acc_1_calib_20.ipynb

    Example #3: 3_EdgeAI_Model_Compile_Run_8b_16b_mixed.ipynb

    Once you have these code examples running, you can read through the code along with the webinar slides and video to understand what each of the code segment is doing.

    Thank you and let us know if you run into any issues.

    Welcome to the AI world with TI Jacinto solutions!

    J7AI_Webinar_DL_Deployment_Jul21.zip