Torchvision version Jetpack 6.0

I installed CUDA enabled pytorch with:
pip3 install --no-cache https://developer.download.nvidia.com/compute/redist/jp/v60/pytorch/torch-2.4.0a0+07cecf4168.nv24.05.14710581-cp310-cp310-linux_aarch64.whl

and get at the end:
ERROR: pip’s dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
torchvision 0.19.0 requires torch==2.4.0, but you have torch 2.4.0a0+07cecf4168.nv24.5 which is incompatible.
Successfully installed torch-2.4.0a0+07cecf4168.nv24.5

When i ran my code i get:
Traceback (most recent call last):
File “/home/spacer/FlareRemoval/Flare7K/test_large.py”, line 3, in
import torchvision
File “/home/spacer/FlareRemoval/venv/lib/python3.10/site-packages/torchvision/init.py”, line 10, in
from torchvision import _meta_registrations, datasets, io, models, ops, transforms, utils # usort:skip
File “/home/spacer/FlareRemoval/venv/lib/python3.10/site-packages/torchvision/_meta_registrations.py”, line 163, in
@torch.library.register_fake(“torchvision::nms”)
AttributeError: module ‘torch.library’ has no attribute ‘register_fake’

It seems a compatibility issue according to the error i got on the pytorch installation but i dont know the corresponding torchvision version needed.

nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2023 NVIDIA Corporation
Built on Tue_Aug_15_22:08:11_PDT_2023
Cuda compilation tools, release 12.2, V12.2.140
Build cuda_12.2.r12.2/compiler.33191640_0

cat /etc/nv_tegra_release

R36 (release), REVISION: 3.0, GCID: 36191598, BOARD: generic, EABI: aarch64, DATE: Mon May 6 17:34:21 UTC 2024

KERNEL_VARIANT: oot

TARGET_USERSPACE_LIB_DIR=nvidia
TARGET_USERSPACE_LIB_DIR_PATH=usr/lib/aarch64-linux-gnu/nvidia

Thanks

Hi @alejandro.serna, presuming that you did a pip3 install torchvision - you need to build torchvision from source (or use the precompiled wheels/docker from jetson-containers)

You can find the instructions for this under the Installation section of this thread:

I haven’t tried PyTorch 2.4 yet, but would try the latest torchvision release with it (0.19)

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