Edge AI model training, quantization, compilation/benchmark & Model Zoo
Our documentation landing pages are the following:
- https://www.ti.com/edgeai : Technology page summarizing TI’s edge AI software/hardware products
- https://github.com/TexasInstruments/edgeai : Landing page for developers to understand overall software and tools offering. Read this before navigating into this repository.
- Our repositories have been restructured : Several repositories are now packaged as components inside this repository.
Please see the release notes
- The subcomponents have detailed documentation. In the browser, navigate into the sub-folders to see detailed documentation. Here is a high level overview.
Category | Tool/Link | Purpose | IS NOT |
---|---|---|---|
List of recommended models and associated information | edgeai-modelzoo | Provides collection of pretrained models | |
Model optimization tools for model training | Model optimization tools | Model optimization tools - Model surgery: Modifies models with minimal loss in accuracy and makes it suitable for TI device (replaces unsupported operators) - Model Pruning/sparsity: Induces sparsity during training – only applicable for specific devices - QAT: Quantization Aware Training to improve accuracy with fixed point quantization |
- Does not support Tensorflow |
Model compilation, accuracy and performance benchmarking | edgeai-benchmark | Bring your own model and compile, benchmark and generate artifacts for deployment on SDK with camera, inference and display (using edgeai-gst-apps) - Comprehends inference pipeline including dataset loading, pre-processing and post-processing - Benchmarking of accuracy and latency with large data sets - Post training quantization - Docker for easy development environment setup |
|
Model training tools | edgeai-torchvision edgeai-mmdetection edgeai-yolox edgeai-mmdetection3d |
Training repositories for various tasks - Provides extensions of popular training repositories (like mmdetection, yolox) with lite version of models |
- Does not support Tensorflow |
Example datasets | edgeai-datasets | Example datasets | |
End-to-End model development; Bring your own data. Integrates several of the above tools. | edgeai-modelmaker (Model Maker) | Integrated, Command line environment for training & compilation - Bring your own data, select a model, perform training and generate artifacts for deployment on SDK - Backend for the GUI tool Model Composer (early availability of features compared to Model Composer). - For novice users, GUI based Model Composer is recommended instead of Model Maker. |
- Does not support Bring Your Own Model workflow |