Seeking Guidance on Real-time Data Stream Update for Continual Learning

[wsy-00]

Dear Avalanche Development Team,

As a beginner in continual learning, my primary goal is to leverage real-time generated monitoring data for continual learning purposes. I possess pretrained models and seek to generate experiences dynamically from new data files rather than having a predetermined list of experiences. I think this may align with online continual learning, yet I haven't found specific implementations related to real-time data stream updates in the documentation or example scripts. I suspect my lack of understanding of the code might be the reason. Could you kindly direct me to relevant code sections or provide guidance on where to find information regarding real-time data stream updates? Any suggestions or pointers would be greatly appreciated. Thank you for your assistance!

Best regards,
Wang