/
Process robotics data with HPCC Systems cluster on Kubernetes

Process robotics data with HPCC Systems cluster on Kubernetes

Owned by Lorraine Chapman
Last updated: Jan 14, 2025 by Hugo Watanuki
2 min read

This project was completed by a student accepted on to the 2021 HPCC Systems Intern Program.

Project Description

This will is a continuing work from last year "Process robotics data with HPCC Systems". The main focus will be on HPCC System cluster on Kubernetes, particularly on Microsoft Azure. The project will adopt existing General Neural Network (GNN) model to local and Azure Kubernetes cluster. Some related code and environment may also need be updated for example, latest ROS, Ubuntu 20.04 and potential new TensorFlow release, etc. The student also will help to identify any necessary change or add-on in HPCC-Platform to support Machine Learning on HPCC System Cloud in both local and public cloud such as Azure, AWS and Google Cloud.

Completion of this project involves:

  • Learning HPCC System ML GNN

  • Learning previous Robotics GNN code

  • Collect or get train data (images) 

  • Load the image data to the cloud

  • Train the model 

By the mid term review we would expect you to have:

  • Load image data to the cloud 

  • Written initial ECL code

  • Trained a model with HPCC Systems GNN with some initial result

Mentor

Mentor: David De Hilster <David.Dehilster@lexisnexisrisk.com>

Backup Mentor: Fortil, Godson  <Godson.Fortil@lexisnexisrisk.com>

Skills needed

  • Docker and Kubernetes

  • Git, CMake

  • Build ROS package

  • HPCC Systems Platform and ECL

  • Python, Unix Bash,

  • Machine Learning, Neural Networks, particularly Convolution Neural Networks (CNN)

  • Kera, TensorFlow

  • Ability to build and test the HPCC system (guidance will be provided).

  • Ability to write test code. Knowledge of ECL is not a requirement since it should be possible to re-use existing code with minimal changes for this purpose. Links are provided below to our ECL training documentation and online courses should you wish to become familiar with the ECL  language.

Deliverables

Midterm

  • Load image data to the cloud 

  • Written initial ECL code

  • Trained a model with HPCC Systems GNN with some initial result

End of project

  • Tuning hyperparameters to improve model training.

  • Documentation

  • A complete github project

Other resources

All pages in this wiki are subject to our site usage guidelines.