• State Resolved
  • Locked Locked
  • Replies 7 replies
  • Subscribers 99 subscribers
  • Views 860 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

SK-TDA4VM: Edge AI Studio Modeling

Vic Chang
Intellectual 350 points

Part Number: SK-TDA4VM
Other Parts Discussed in Thread: EDGE-AI-STUDIO

Hi


I have a question about Model Composer, Model Composer, Model Composer.

My understanding of EDGE-AI-STUDIO is that TI provides a method similar to online model training at

Edge AI Studio (ti.com)

If I have my own trained models, can I use the TI platform for the following purposes?

1. Analysis and optimization/model compression/quantization acceleration?

2. Use non-default Image Classification/Feature Extraction/Object Detection/Semantic Segmentation/Human Pose Estimation/3D Object Detection models?

Can I design and create my own models, possibly not following a conventional architecture, and run them on the platform?

  • 0
    TI__Mastermind 42135 points

    Hi,

    The engineer is OoO for regional holidays until 11/20 so please expect delay in response.
    Apologies for the inconvenience and thank you for your patience.

    Regards,
    Parth

  • 0 in reply to Vic Chang
    TI__Mastermind 42135 points

    Thank you for your patience.

    Regards,
    Parth

  • 0 in reply to Parth Nagpal
    TI__Mastermind 24041 points

    Hi Vic,

    As you have queries are related specifically to model composer tool let me answer them.

    Vic Chang said:
    My understanding of EDGE-AI-STUDIO is that TI provides a method similar to online model training at

    Yes, you can train the model by feeding images for different supported tasks like object detection and classification.

    Vic Chang said:
    1. Analysis and optimization/model compression/quantization acceleration?

    You will get different graph level analysis info on epoch stage accuracy of current model and many more when training is on going. You can opt for supported quantization rage option 8/16 bits. Not sure what you mean by compression here, can you elaborate more on this term ?

    Vic Chang said:
    2. Use non-default Image Classification/Feature Extraction/Object Detection/Semantic Segmentation/Human Pose Estimation/3D Object Detection models?
    Vic Chang said:
    Can I design and create my own models, possibly not following a conventional architecture, and run them on the platform?

    Per my understanding currently custom models are not supported as part of this tool.

    However you can look at our edgeai tild tools repos for custom model training here : https://github.com/TexasInstruments/edgeai-tidl-tools

  • 0 in reply to Pratik Kedar
    Intellectual 350 points

    Hi 

    The things I'm discussing are not problems, so there's no need to feel pressured or worried. The main focus is on understanding the situation and feasibility of this platform.

    Q1
    Model Compression:
    Quantization:
    Basically, it involves the conversion of floating point to integer for compression and acceleration.

    Other Techniques:
    Pruning, Distillation, Structural Optimization.

    Q2
    Custom Models:
    In line with the Q1 question, other techniques involve modifications to the model structure (model compression). Q2 is more like a discussion on the portability of models.

    If I have a trained model, can it be directly transformed for hardware acceleration?
    Not for Classification/Feature Extraction/Object Detection/Semantic Segmentation/Human Pose Estimation/3D Object Detection models.
    It's about a generic model structure.

  • +1 in reply to Vic Chang
    TI__Mastermind 24041 points

    Hi,

    Vic Chang said:
    Q1
    Model Compression:
    Quantization:
    Basically, it involves the conversion of floating point to integer for compression and acceleration.

    Thanks for clarification on term Compression yes quantization support is available and user can opt for 8/16 bit range.

    Vic Chang said:
    If I have a trained model, can it be directly transformed for hardware acceleration?

    On EdgeAI model composer tool side this support is not available, however as mentioned above you can use our edgeai tidl tools repos to achieve custom model compilation.

    Pratik Kedar said:
    Can I design and create my own models, possibly not following a conventional architecture, and run them on the platform?

    Per my understanding currently custom models are not supported as part of this tool.

    However you can look at our edgeai tild tools repos for custom model training here : https://github.com/TexasInstruments/edgeai-tidl-tools

    Hope this answers your queries, I would recommend to please read through the edgeai tidl tools link shared above to understand custom trained model compilation.

  • 0 in reply to Pratik Kedar
    Intellectual 350 points

    Hi 

    Thank you!

    I'll give it a try.