Incompatible with Blackwell series chip

[Robert-Jia00129]

Already has a solution. Will be sending a PR for the fix. Error below.

(my) root@atas-fpg67-02:/workspace# python scripts/run_demo.py --model_dir weights/23-36-37/model_best_bp2_serialize.pth --left_file demo_data/left.png --right_file demo_data/right.png --intrinsic_file demo_data/K.txt --out_dir output/ --remove_invisible 0 --denoise_cloud 1  --scale 1 --get_pc 1 --valid_iters 8 --max_disp 192 --zfar 100
/opt/conda/envs/my/lib/python3.12/site-packages/torch/cuda/__init__.py:235: UserWarning: 
NVIDIA GeForce RTX 5060 Ti with CUDA capability sm_120 is not compatible with the current PyTorch installation.
The current PyTorch install supports CUDA capabilities sm_50 sm_60 sm_70 sm_75 sm_80 sm_86 sm_90.
If you want to use the NVIDIA GeForce RTX 5060 Ti GPU with PyTorch, please check the instructions at https://pytorch.org/get-started/locally/

  warnings.warn(
args:
{'corr_levels': 2, 'corr_radius': 4, 'hidden_dims': [128], 'low_memory': 0, 'max_disp': 192, 'mixed_precision': True, 'n_downsample': 2, 'n_gru_layers': 1, 'slow_fast_gru': False, 'valid_iters': 8, 'vit_size': 'vitl', 'model_dir': 'weights/23-36-37/model_best_bp2_serialize.pth', 'left_file': 'demo_data/left.png', 'right_file': 'demo_data/right.png', 'intrinsic_file': 'demo_data/K.txt', 'out_dir': 'output/', 'remove_invisible': 0, 'denoise_cloud': 1, 'denoise_nb_points': 30, 'denoise_radius': 0.03, 'scale': 1.0, 'hiera': 0, 'get_pc': 1, 'zfar': 100.0}
/opt/conda/envs/my/lib/python3.12/site-packages/torch/cuda/__init__.py:235: UserWarning: 
NVIDIA GeForce RTX 5060 Ti with CUDA capability sm_120 is not compatible with the current PyTorch installation.
The current PyTorch install supports CUDA capabilities sm_50 sm_60 sm_70 sm_75 sm_80 sm_86 sm_90.
If you want to use the NVIDIA GeForce RTX 5060 Ti GPU with PyTorch, please check the instructions at https://pytorch.org/get-started/locally/

  warnings.warn(
/workspace/scripts/run_demo.py:64: DeprecationWarning: Starting with ImageIO v3 the behavior of this function will switch to that of iio.v3.imread. To keep the current behavior (and make this warning disappear) use `import imageio.v2 as imageio` or call `imageio.v2.imread` directly.
  img0 = imageio.imread(args.left_file)
/workspace/scripts/run_demo.py:65: DeprecationWarning: Starting with ImageIO v3 the behavior of this function will switch to that of iio.v3.imread. To keep the current behavior (and make this warning disappear) use `import imageio.v2 as imageio` or call `imageio.v2.imread` directly.
  img1 = imageio.imread(args.right_file)
img0: (540, 960, 3)
Traceback (most recent call last):
  File "/workspace/scripts/run_demo.py", line 82, in <module>
    img0 = torch.as_tensor(img0).cuda().float()[None].permute(0,3,1,2)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
RuntimeError: CUDA error: no kernel image is available for execution on the device
CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.
For debugging consider passing CUDA_LAUNCH_BLOCKING=1
Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions.

[Robert-Jia00129]

#46