Copy GPT-2 artifacts to Raspberry Pi
Run on your development machine to copy your model artifacts over to the target Pi. Ensure you substitute your Raspberry Pi username and Raspberry Pi IP address:
cd ~/executorch_on_arm
ssh <user>@<pi> "mkdir -p ~/executorch_on_arm/module-1-transformer-inference-with-xnnpack/artifacts"
scp -r module-1-transformer-inference-with-xnnpack/artifacts/gpt2 <user>@<pi>:~/executorch_on_arm/module-1-transformer-inference-with-xnnpack/artifacts/
This copies the GPT-2 spec, .pte artifacts, and metadata to the Pi. It may take a few minutes because the files are large.
Course complete
Nice work
You completed Optimize and deploy AI inference with ExecuTorch on Arm.