Can we load multiple Deep Learning models on a Jetson tx2

Autonomous Machines Jetson & Embedded Systems Jetson TX2
1

Hi,

Can we load two deep learning models using Python on Jetson tx2? For example does it have sufficient memory to load two Inception ResNet models in memory such that one does detection and the other does classification?

Just curious to know if someone managed this on Tx2? Is there a smarter way to do this otherwise pls?

Thanks
Abhishek S

2

Hi,

Please noticed that we don’t provide an API for python on Jetson.
But you can do this via some third-party interface, ex. TensorFlow or Caffe.

Here is an quite approach to test your requirement:

1. Compile TensorRT

cp -r /usr/src/tensorrt/ .
cd tensorrt/samples/trtexec/
make
cd ../../bin/

2. Run this command in multiple console.

./giexec --deploy=/path/to/prototxt --output=/name/of/output

Thanks.