Awesome
Edge AI Software And Development Tools
<hr>Release Notes
- [2024-September] 10.0 release has been done. SDKs, edgeai-tidl-tools and edgeai-tensorlab has been updated.
Further details are in the Release Notes.
Also see the SDKs release notes, edgeai-tidl-tools release notes and edgeai-tensorlab release notes
<hr>Notice
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
- Our repositories have been restructured : Please navigate to the tables below to understand how several repositories are now packaged inside edgeai-tensorlab
Introduction
Embedded inference of Deep Learning models is quite challenging - due to high compute requirements. TI’s Edge AI comprehensive software product help to optimize and accelerate inference on TI’s embedded devices. It supports heterogeneous execution of DNNs across cortex-A based MPUs, TI’s latest generation C7x DSP and DNN accelerator (MMA).
TI's Edge AI solution simplifies the whole product life cycle of DNN development and deployment by providing a rich set of tools and optimized libraries.
Overview
The figure below provides a high level summary of the relevant tools:<br><img src="assets/workblocks_tools_software.png" width="800">
<hr>Details of various tools
The table below provides detailed explanation of each of the tools:
| Category | Tool/Link | Purpose | IS NOT |
|---|---|---|---|
| Inference (and compilation) Tools | edgeai-tidl-tools | To get familiar with model compilation and inference flow<br>- Post training quantization<br>- Benchmark latency with out of box example models (10+)<br>- Compile user / custom model for deployment<br>- Inference of compiled models on X86_PC or TI SOC using file base input and output<br>- Docker for easy development environment setup | - Does not support benchmarking accuracy of models using TIDL with standard datasets, for e.g. - accuracy benchmarking using MS COCO dataset for object detection models. Please refer to edgeai-benchmark for the same.<br>- Does not support Camera, Display and inference based end-to-end pipeline development. Please refer Edge AI SDK for such usage |
| Model Selection Tool | Edge AI Studio: Model Selection Tool | Understand performance statistics of models such as FPS, Latency, Accuracy & DDR bandwidth. Find the model that best meets your performance and accuracy goals on TI Processor from TI Model Zoo. | |
| Integrated environment for training and compilation | Edge AI Studio: Model Analyzer | Browser based environment to allow model evaluation with TI EVM farm<br>- Allow model evaluation without and software/hardware setup at user end<br>- User can reserve EVM from TI EVM farm and perform model evaluation using jupyter notebook<br>- Model selection tool: To provide suitable model architectures for TI devices | - Does not support Camera, Display and inference based end-to-end pipeline development. Please refer Edge AI SDK for such usage |
| ditto | Edge AI Studio: Model Composer | GUI based Integrated environment for data set capture, annotation, training, compilation with connectivity to TI development board<br>- Bring/Capture your own data, annotate, select a model, perform training and generate artifacts for deployment on SDK<br>- Live preview for quick feedback | - Does not support Bring Your Own Model workflow |
| Edge AI Software Development Kit | Devices & SDKs | SDK to develop end-to-end AI pipeline with camera, inference and display<br>- Different inference runtime: TFLiteRT, ONNXRT, NEO AI DLR, TIDL-RT<br>- Framework: openVX, gstreamer<br>- Device drivers: Camera, display, networking<br>- OS: Linux, RTOS<br>- May other software modules: codecs, OpenCV,… |
| Category | Tool/Link | Purpose | IS NOT |
|---|---|---|---|
| Model Zoo, Model training, compilation/benchmark & associated tools | edgeai-tensorlab | To provide model training software, collection of pretrained models and documemtation and compilation/benchmark scripts. Includes edgeai-modelzoo, edgeai-benchmark, edgeai-modeloptimization, edgeai-modelmaker, edgeai-torchvision, edgeai-mmdetection and such repositories. |
Workflows
Bring your own model (BYOM) workflow:<br><img src="assets/workflow_bring_your_own_model.png" width="600">
Train your own model (TYOM) workflow:<br><img src="assets/workflow_train_your_own_model.png" width="600">
Bring your own data (BYOD) workflow:<br><img src="assets/workflow_bring_your_own_data.png" width="600">
<hr>Tech Reports
Technical documentation can be found in the documentation of each repository. Here we have a collection of technical reports & tutorials that give high level overview on various topics.
<hr>Publications
- Read some of our Technical publications
Issue Trackers
Issue tracker for Edge AI Studio is listed in its landing page.
Issue tracker for TIDL: Please include the tag TIDL (as you create a new issue, there is a space to enter tags, at the bottom of the page).
Issue tracker for edge AI SDK Please include the tag EDGEAI (as you create a new issue, there is a space to enter tags, at the bottom of the page).
Issue tracker for ModelZoo, Model Benchmark & Deep Neural Network Training Software: Please include the tag MODELZOO (as you create a new issue, there is a space to enter tags, at the bottom of the page).
<hr>License
Please see the LICENSE file for more information about the license under which this landing repository is made available. The LICENSE file of each repository is inside that repository.