WHAM version conflicts with DPVO about wham

[hanlinniu]

Currently, I have this error:

2024-06-15 21:10:17.769 | INFO | main::209 - GPU name -> NVIDIA GeForce RTX 3080 Ti Laptop GPU
2024-06-15 21:10:17.769 | INFO | main::210 - GPU feat -> CudaDeviceProperties(name='NVIDIA GeForce RTX 3080 Ti Laptop GPU', major=8, minor=6, total_memory=16125MB, multi_processor_count=58)
2024-06-15 21:10:18.809 | INFO | lib.models:build_network:36 - => loaded checkpoint 'checkpoints/wham_vit_bedlam_w_3dpw.pth.tar'
load checkpoint from local path: /home/hanlin/WHAM/checkpoints/vitpose-h-multi-coco.pth
Load backbone weight: /home/hanlin/WHAM/checkpoints/hmr2a.ckpt
Traceback (most recent call last):
File "/home/hanlin/WHAM/demo.py", line 224, in
run(cfg,
File "/home/hanlin/WHAM/demo.py", line 65, in run
detector.track(img, fps, length)
File "/home/hanlin/WHAM/lib/models/preproc/detector.py", line 82, in track
bboxes = self.bbox_model.predict(
File "/home/hanlin/anaconda3/envs/wham/lib/python3.9/site-packages/ultralytics/engine/model.py", line 446, in predict
self.predictor.setup_model(model=self.model, verbose=is_cli)
File "/home/hanlin/anaconda3/envs/wham/lib/python3.9/site-packages/ultralytics/engine/predictor.py", line 303, in setup_model
self.model = AutoBackend(
File "/home/hanlin/anaconda3/envs/wham/lib/python3.9/site-packages/torch/autograd/grad_mode.py", line 27, in decorate_context
return func(*args, **kwargs)
File "/home/hanlin/anaconda3/envs/wham/lib/python3.9/site-packages/ultralytics/nn/autobackend.py", line 144, in init
model = model.fuse(verbose=verbose)
File "/home/hanlin/anaconda3/envs/wham/lib/python3.9/site-packages/ultralytics/nn/tasks.py", line 184, in fuse
m.conv = fuse_conv_and_bn(m.conv, m.bn) # update conv
File "/home/hanlin/anaconda3/envs/wham/lib/python3.9/site-packages/ultralytics/utils/torch_utils.py", line 205, in fuse_conv_and_bn
fusedconv.weight.copy(torch.mm(w_bn, w_conv).view(fusedconv.weight.shape))
RuntimeError: CUDA error: CUBLAS_STATUS_INVALID_VALUE when calling cublasSgemm( handle, opa, opb, m, n, k, &alpha, a, lda, b, ldb, &beta, c, ldc)

I am using cuda 11.3, cudnn 8.2

[ChrisChenYh]

runing unset LD_LIBRARY_PATH may be helpful

[ChrisChenYh]

And how you solve the version conflicts? First install DPVO, and then modify the environments of WHAM?