tango-23.11

[ML-TANGO]

TANGO 2023 11 Release

This release includes several enhancements and updates on following functions:

• Project Manager: project_manager,
• Base Model Selector: bms,
• AutoNN : autonn_yoloe (for detection task), autonn_resnet (newly added for classification task)
• Code Generation: code_gen (formerly known as deploy_codegen),
• Target Deployment: cloud_deploy, kube_deploy, ondevice_deploy
  • formerly single deploy_target container did the whole deployment task on different targets, now we have individual deployment container for each target

To improve the efficiency of the autonn container, which typically requires several days for training even on high-performance GPUs, we have implemented a manual workflow step-forward functionality within the project manager. This functionality has been tested and validated through two stages:

• bms + autonn_yoloe or autonn_resnet stage: This stage utilizes the dataset and target configurations within the project manager to select the appropriate base model. The selected base model undergoes fine-tuning, generating a trained model and associated codes for the subsequent stage.
• code_gen + *_deploy stage: Building upon the trained model and generated codes from the previous stage, this stage prepares executable neural network codes for deployment on the specified target, as configured in the project_manager container. Please note that the current release includes the addition of K8s and cloud (ex. Google Cloud Platform) target deployment-related code, although it is still under going.

Notes on the current release:

• We have tested K8s target deployment-related code, which has been developed and built.
• The project manager in this release is based on Vue.js based front-end.

BMS and AutoNN

This release includes the bms (Base Model Selector) container, developed by ETRI, which serves as a simple test for BMS member container role. The BMS container selects the base model from the Yolo v7 series or Resnet series and suitable batch size for training based on target type information such as ondevice (PC, Andorid Device, Embedded Board), K8s or cloud, specified within the project configuration step of the project manager. The selected base model is utilized in the AutoNN containers for fine-tuning with the dataset, also specifed within the project configuration step. Additionally, the AutoNN yoloe and resnet containers (implemented in autonn\YoloE and autonn\ResNet folders) hvae been included for testing the AutoNN member container role within the TANGO project workflow (pipeline).

Code Generation and Deployment

We have made changes to the source structure related to deployment codes compared to the previous release. The updates are as follows:

• The codes responsible for generating the executable neural network code have been moved to the deploy_codegen/optimize_codegen folder.
• The codes for deploying the executable neural network have been relocated to the deploy_target folder.

The deploy_target folder now includes sub-folders based on the deployment target:

• cloud: codes for deployment to cloud environments
• k8s: codes for deployment to Kubernetes clusters (Please note that the k8s deployment functionality is not operational in this release but will be addressed in future updates.)
• ondevice: codes for deployment to on-device platforms such as PC, android phones, or embedded devices

Notes on deployment:

• The codes in the deploy_target/k8s folder can be used to build a Docker image, but integration testing is still ongoing. Therefore, in this release, K8s deployment is not functioning properly.
• In this release, code_gen container supports TensorRT, Apache TVM, PyTorch and [ACL] (https://github.com/ARM-software/ComputeLibrary). Support for RKNN . Support for rknn is still in progress.

To Do:

---

This discussion was created from the release tango-23.11.