runtime error

Exit code: 1. Reason: ting uninstall: triton Found existing installation: triton 2.3.1 Uninstalling triton-2.3.1: Successfully uninstalled triton-2.3.1 Attempting uninstall: nvidia-nccl-cu12 Found existing installation: nvidia-nccl-cu12 2.20.5 Uninstalling nvidia-nccl-cu12-2.20.5: Successfully uninstalled nvidia-nccl-cu12-2.20.5 Attempting uninstall: torch Found existing installation: torch 2.3.1 Uninstalling torch-2.3.1: Successfully uninstalled torch-2.3.1 Successfully installed accelerate-0.21.0 bitsandbytes-0.43.3 einops-0.6.1 einops-exts-0.0.4 joblib-1.4.2 latex2mathml-3.77.0 llava-1.2.2.post1 markdown2-2.5.0 nvidia-nccl-cu12-2.18.1 peft-0.12.0 regex-2024.7.24 safetensors-0.4.4 scikit-learn-1.2.2 scipy-1.14.1 sentencepiece-0.1.99 shortuuid-1.0.13 svgwrite-1.4.3 threadpoolctl-3.5.0 timm-0.6.13 tokenizers-0.15.1 torch-2.1.2 torchvision-0.16.2 transformers-4.37.2 triton-2.1.0 wavedrom-2.0.3.post3 [notice] A new release of pip available: 22.3.1 -> 24.2 [notice] To update, run: pip install --upgrade pip /usr/local/lib/python3.10/site-packages/torch/cuda/__init__.py:611: UserWarning: Can't initialize NVML warnings.warn("Can't initialize NVML") The installed version of bitsandbytes was compiled without GPU support. 8-bit optimizers, 8-bit multiplication, and GPU quantization are unavailable. `flash-attention` package not found, consider installing for better performance: No module named 'flash_attn'. Current `flash-attenton` does not support `window_size`. Either upgrade or use `attn_implementation='eager'`. ↑ Those bitsandbytes warnings are expected on ZeroGPU ↑ Traceback (most recent call last): File "/home/user/app/app.py", line 15, in <module> import spaces File "/usr/local/lib/python3.10/site-packages/spaces/__init__.py", line 19, in <module> from .zero.decorator import GPU File "/usr/local/lib/python3.10/site-packages/spaces/zero/__init__.py", line 15, in <module> raise RuntimeError( RuntimeError: CUDA has been initialized before importing the `spaces` package

Container logs: