Simulataneous PyTorch and jtop logging

I want to log the training and inference process of a PyTorch neural network, including the power consumption with jtop. The limitation of using wandb.ai is that it requires either (a) network connection or (b) a self hosted docker container both of which disturb the power measurements.

Any recommendations on how I could go about this?

Hi,

Please refer to the example of GitHub - rbonghi/jetson_stats: 📊 Simple package for monitoring and control your NVIDIA Jetson [Orin, Xavier, Nano, TX] series

Thanks

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.