Unable to download torchvision whl

Robotics & Edge Computing Jetson & Embedded Systems Jetson Orin NX
,
1

Continuing the discussion from PyTorch and Torvision version issue: RuntimeError: operator torchvision::nms does not exist:

when trying to access above URL to download torchvision, the browser shows “The site cannot be reached”.

I have tried multiple networks but facing same issues

2

*** Please note that this reply is generated by LLM automatically ***

I’d be happy to help you troubleshoot the issue!

It appears that you’re trying to download Torchvision, but the browser is showing “The site cannot be reached.” This error message typically indicates a network connectivity issue or a problem with the website itself.

Since you’ve tried multiple networks and are still facing the same issue, let’s explore a few potential solutions:

  1. Check the URL: Double-check that the URL you’re trying to access is correct. Ensure that there are no typos or incorrect characters in the URL.
  2. Network connectivity: Verify that your internet connection is stable and working properly. Try accessing other websites to confirm that the issue is specific to the Torchvision download URL.
  3. Browser cache and cookies: Clear your browser’s cache and cookies, then try accessing the URL again. This can sometimes resolve issues related to corrupted cache or cookies.
  4. Alternative download methods: Instead of using a browser, try using a command-line tool like wget or curl to download the Torchvision package. This can help bypass any browser-specific issues.
  5. Check for mirror sites: If the official Torchvision website is down or inaccessible, you can try searching for mirror sites or alternative download locations.
  6. PyTorch NGC container: As an alternative, you can try using the PyTorch NGC container, which includes pre-installed PyTorch and Torchvision packages. This might save you the trouble of downloading and installing Torchvision separately.

To use the PyTorch NGC container, you can follow these steps:

a. Pull the PyTorch NGC container using docker pull nvcr.io/nvidia/pytorch:20.03-py3 (or the version that matches your PyTorch version).
b. Run the container using docker run --gpus all -it --rm nvcr.io/nvidia/pytorch:20.03-py3 (or the version that matches your PyTorch version).
c. Inside the container, you can verify that PyTorch and Torchvision are installed by running python -c "import torch; import torchvision".

If none of these solutions work, please provide more details about your environment, such as your operating system, PyTorch version, and network setup. This will help me better understand the issue and provide more tailored assistance.

3

With our current requirement I cannot use docker, also I have tried wget and validated the URL. but still getting same error.

pip install http://jetson.webredirect.org/jp6/cu126/+f/5f9/67f920de3953f/torchvision-0.20.0-cp310-cp310-linux_aarch64.whl#sha256=5f967f920de3953f2a39d95154b1feffd5ccc06b4589e51540dc070021a9adb9
Looking in indexes: https://pypi.org/simple, https://pypi.ngc.nvidia.com
Collecting torchvision==0.20.0
  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 0xffffa4d394b0>: Failed to establish a new connection: [Errno 113] No route to host')': /jp6/cu126/+f/5f9/67f920de3953f/torchvision-0.20.0-cp310-cp310-linux_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 0xffffa4d39780>: Failed to establish a new connection: [Errno 113] No route to host')': /jp6/cu126/+f/5f9/67f920de3953f/torchvision-0.20.0-cp310-cp310-linux_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 0xffffa4d39930>: Failed to establish a new connection: [Errno 113] No route to host')': /jp6/cu126/+f/5f9/67f920de3953f/torchvision-0.20.0-cp310-cp310-linux_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 0xffffa4d39ae0>: Failed to establish a new connection: [Errno 113] No route to host')': /jp6/cu126/+f/5f9/67f920de3953f/torchvision-0.20.0-cp310-cp310-linux_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 0xffffa4d39c90>: Failed to establish a new connection: [Errno 113] No route to host')': /jp6/cu126/+f/5f9/67f920de3953f/torchvision-0.20.0-cp310-cp310-linux_aarch64.whl
ERROR: Could not install packages due to an OSError: HTTPSConnectionPool(host='pypi.jetson-ai-lab.dev', port=443): Max retries exceeded with url: /jp6/cu126/+f/5f9/67f920de3953f/torchvision-0.20.0-cp310-cp310-linux_aarch64.whl (Caused by NewConnectionError('<pip._vendor.urllib3.connection.HTTPSConnection object at 0xffffa4d39db0>: Failed to establish a new connection: [Errno 113] No route to host'))
4

@AastaLLL Could you please help with this issue?

5

I am unable to find .whl for torchvision==0.20.0 on jp6/cu126 index

7

I am facing the same issue. It seems there is no .whl for torchvision

8

link doesn’t exist anymore !! Need it too !

9

pypi.jetson-ai-lab.dev is down. We are improving security and performance. Meanwhile please use pypi.jetson-ai-lab.io by setting the enviroment variable INDEX_HOST=jetson-ai-lab.io in .env file.

You can also refer to Jetson-ai-lab server down? - Jetson & Embedded Systems / Jetson Orin Nano - NVIDIA Developer Forums

11

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