Fix the official environment errors/incompatible

# Step 1: Install Anaconda
bash Anaconda3-2024.10-1-Linux-x86_64.sh

# Step 2: Create conda environment from YAML
conda env create -f dinov2-extras.yaml

# Step 3: Activate the environment
conda activate dinov2-extras

# Step 4: Install necessary packages
pip install -U openmim
mim install mmcv-full
pip install mmsegmentation==0.30.0 xformers==0.0.18

If you get a warning saying that PyTorch CUDA version is not matching, do the following:

# Uninstall current PyTorch packages
pip uninstall torch torchvision torchaudio -y

# Reinstall with the correct CUDA version (11.8)
pip install torch==2.0.0+cu118 torchvision==0.15.0+cu118 torchaudio==2.0.0 --index-url https://download.pytorch.org/whl/cu118

1. Download DINOv2 from GitHub and unzip it.
2. Navigate to the folder: ./dinov2 (⚠️ Not ./dinov2/dinov2)
3. Follow the steps in Installing Step One by One above.
4. Activate the conda environment:

   conda activate dinov2-extras

5. Run the script:

   python run.py --input <picture folder path> --output <output path>

6. Set the desired backbone and head in the script or via command line.
7. The run.py script will automatically:
   • Load all .png images in <picture folder path>
   • Process them into depth maps
   • Save the results in <output path> with the same format and filename

Happy predicting! 😄