TAO-Deploy 5 on Jetson

• Hardware: AGX Orin
• Network Type: Any
• TAO Version: 5.0.0
• How to reproduce the issue ? (This is for errors. Please share the command line and the detailed log here.)

Hi.

I am trying to use TAO-Deploy to export a model to .engine format for a Jetson device. As mentioned in the TAO Deploy Github, you need to run the following commands when using tao-deploy on aarch64 devices.

sudo docker run -it --rm --net=host --runtime nvidia nvcr.io/nvidia/l4t-tensorrt:8.5.2.2-devel /bin/bash

apt update
apt install libopenmpi-dev
pip install nvidia_tao_deploy==5.0.0.423.dev0
pip install https://urm.nvidia.com/artifactory/sw-eff-pypi/nvidia-eff-tao-encryption/0.1.7/nvidia_eff_tao_encryption-0.1.7-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl
pip install https://urm.nvidia.com/artifactory/sw-eff-pypi/nvidia-eff/0.6.2/nvidia_eff-0.6.2-py38-none-manylinux_2_17_aarch64.manylinux2014_aarch64.whl

However, I am unable to download the nvidia-eff-tao-encryption and nvidia_eff wheels, logs below.

pip install https://urm.nvidia.com/artifactory/sw-eff-pypi/nvidia-eff-tao-encryption/0.1.7/nvidia_eff_tao_encryption-0.1.7-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl
Collecting nvidia-eff-tao-encryption==0.1.7
  WARNING: Retrying (Retry(total=4, connect=None, read=None, redirect=None, status=None)) after connection broken by 'NewConnectionError('<pip._vendor.urllib3.connection.HTTPSConnection object at 0xffff854a4b20>: Failed to establish a new connection: [Errno -5] No address associated with hostname')': /artifactory/sw-eff-pypi/nvidia-eff-tao-encryption/0.1.7/nvidia_eff_tao_encryption-0.1.7-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl
  WARNING: Retrying (Retry(total=3, connect=None, read=None, redirect=None, status=None)) after connection broken by 'NewConnectionError('<pip._vendor.urllib3.connection.HTTPSConnection object at 0xffff854a4e20>: Failed to establish a new connection: [Errno -5] No address associated with hostname')': /artifactory/sw-eff-pypi/nvidia-eff-tao-encryption/0.1.7/nvidia_eff_tao_encryption-0.1.7-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl
  WARNING: Retrying (Retry(total=2, connect=None, read=None, redirect=None, status=None)) after connection broken by 'NewConnectionError('<pip._vendor.urllib3.connection.HTTPSConnection object at 0xffff854c20d0>: Failed to establish a new connection: [Errno -5] No address associated with hostname')': /artifactory/sw-eff-pypi/nvidia-eff-tao-encryption/0.1.7/nvidia_eff_tao_encryption-0.1.7-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl
  WARNING: Retrying (Retry(total=1, connect=None, read=None, redirect=None, status=None)) after connection broken by 'NewConnectionError('<pip._vendor.urllib3.connection.HTTPSConnection object at 0xffff854c22e0>: Failed to establish a new connection: [Errno -5] No address associated with hostname')': /artifactory/sw-eff-pypi/nvidia-eff-tao-encryption/0.1.7/nvidia_eff_tao_encryption-0.1.7-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl
  WARNING: Retrying (Retry(total=0, connect=None, read=None, redirect=None, status=None)) after connection broken by 'NewConnectionError('<pip._vendor.urllib3.connection.HTTPSConnection object at 0xffff854c24c0>: Failed to establish a new connection: [Errno -5] No address associated with hostname')': /artifactory/sw-eff-pypi/nvidia-eff-tao-encryption/0.1.7/nvidia_eff_tao_encryption-0.1.7-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl
ERROR: Could not install packages due to an OSError: HTTPSConnectionPool(host='urm.nvidia.com', port=443): Max retries exceeded with url: /artifactory/sw-eff-pypi/nvidia-eff-tao-encryption/0.1.7/nvidia_eff_tao_encryption-0.1.7-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl (Caused by NewConnectionError('<pip._vendor.urllib3.connection.HTTPSConnection object at 0xffff854c26a0>: Failed to establish a new connection: [Errno -5] No address associated with hostname'))

Any assistance will be appreciated.

Kind regards,
Nicholas

Please try to install with attachment.
nvidia_eff-0.6.2-py38-none-manylinux_2_17_aarch64.manylinux2014_aarch64.whl (700.6 KB)
nvidia_eff_tao_encryption-0.1.7-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl (2.3 MB)

To narrow down, can you run below commands successfully?

$ wget
https://urm.nvidia.com/artifactory/sw-eff-pypi/nvidia-eff-tao-encryption/0.1.7/nvidia_eff_tao_encryption-0.1.7-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl

$ wget
https://urm.nvidia.com/artifactory/sw-eff-pypi/nvidia-eff/0.6.2/nvidia_eff-0.6.2-py38-none-manylinux_2_17_aarch64.manylinux2014_aarch64.whl

Hi Morganh.

Thanks for the whl files, which installed with no issue. I was able to export the model successfully.

When running the wget command I get the following output:

$ wget https://urm.nvidia.com/artifactory/sw-eff-pypi/nvidia-eff-tao-encryption/0.1.7/nvidia_eff_tao_encryption-0.1.7-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl
--2023-10-10 11:37:01--  https://urm.nvidia.com/artifactory/sw-eff-pypi/nvidia-eff-tao-encryption/0.1.7/nvidia_eff_tao_encryption-0.1.7-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl
Resolving urm.nvidia.com (urm.nvidia.com)... failed: No address associated with hostname.
wget: unable to resolve host address ‘urm.nvidia.com’

$ wget https://urm.nvidia.com/artifactory/sw-eff-pypi/nvidia-eff/0.6.2/nvidia_eff-0.6.2-py38-none-manylinux_2_17_aarch64.manylinux2014_aarch64.whl
--2023-10-10 12:55:24--  https://urm.nvidia.com/artifactory/sw-eff-pypi/nvidia-eff/0.6.2/nvidia_eff-0.6.2-py38-none-manylinux_2_17_aarch64.manylinux2014_aarch64.whl
Resolving urm.nvidia.com (urm.nvidia.com)... failed: No address associated with hostname.
wget: unable to resolve host address ‘urm.nvidia.com’

Kind regards,
Nicholas

Thanks for the info.

Already update https://github.com/NVIDIA/tao_deploy/blob/main/README.md#JetsonDevices.

pip install https://files.pythonhosted.org/packages/f7/7a/ac2e37588fe552b49d8807215b7de224eef60a495391fdacc5fa13732d11/nvidia_eff_tao_encryption-0.1.7-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl
pip install https://files.pythonhosted.org/packages/0d/05/6caf40aefc7ac44708b2dcd5403870181acc1ecdd93fa822370d10cc49f3/nvidia_eff-0.6.2-py38-none-manylinux_2_17_aarch64.manylinux2014_aarch64.whl

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