Extract embeddings/activations from Jetson

  • Is there an easy way in which I can extract intermediate embeddings / activations from a Jetson model, from a chosen layer?

Hi,

Could you share more information about your use case?
Which frameworks do you use for inference?

For TensorRT, you can mark a layer as output to get the intermediate results.
https://docs.nvidia.com/deeplearning/tensorrt/api/python_api/infer/Graph/Network.html?highlight=output#tensorrt.INetworkDefinition.mark_output

Thanks.

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