Beta value for binned Visium HD data & error with slice indices

[Demond-dev]

Hi,

I'm attempting to use SpaGCN on Visium HD data that I've binned. I was wondering what the ideal beta value to use would be? After binning the 2micron bins the anndata object has a calculated spot size diameter. Would it be appropriate to use that value or the 2 micron bin size for performing clustering?

Also I'm running into an error with running clustering with histology where slice indices are not being returned as integers.

# Read in full resolution image

img = cv2.imread(source_image_path)

# Calculate adjacent matrix
adj = spg.calculate_adj_matrix(
    x = x_pixel,
    y = y_pixel,
    x_pixel = x_pixel,
    y_pixel = y_pixel,
    image = img,
    beta = 29.367863437955762,
    alpha = 1,
    histology = True,
)

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
Cell In[16], [line 2](vscode-notebook-cell:?execution_count=16&line=2)
      [1](vscode-notebook-cell:?execution_count=16&line=1) # Calculate adjacent matrix
----> [2](vscode-notebook-cell:?execution_count=16&line=2) adj = spg.calculate_adj_matrix(
      [3](vscode-notebook-cell:?execution_count=16&line=3)     x = x_pixel,
      [4](vscode-notebook-cell:?execution_count=16&line=4)     y = y_pixel,
      [5](vscode-notebook-cell:?execution_count=16&line=5)     x_pixel = x_pixel,
      [6](vscode-notebook-cell:?execution_count=16&line=6)     y_pixel = y_pixel,
      [7](vscode-notebook-cell:?execution_count=16&line=7)     image = img,
      [8](vscode-notebook-cell:?execution_count=16&line=8)     beta = 29.367863437955762,
      [9](vscode-notebook-cell:?execution_count=16&line=9)     alpha = 1,
     [10](vscode-notebook-cell:?execution_count=16&line=10)     histology = True,
     [11](vscode-notebook-cell:?execution_count=16&line=11) )

File /data/williamsd7/conda/envs/spatialstuff/lib/python3.12/site-packages/SpaGCN/calculate_adj.py:57, in calculate_adj_matrix(x, y, x_pixel, y_pixel, image, beta, alpha, histology)
     [55](https://vscode-remote+ssh-002dremote-002bcn4248.vscode-resource.vscode-cdn.net/data/williamsd7/conda/envs/spatialstuff/lib/python3.12/site-packages/SpaGCN/calculate_adj.py:55) 	max_x=image.shape[0]
     [56](https://vscode-remote+ssh-002dremote-002bcn4248.vscode-resource.vscode-cdn.net/data/williamsd7/conda/envs/spatialstuff/lib/python3.12/site-packages/SpaGCN/calculate_adj.py:56) 	max_y=image.shape[1]
---> [57](https://vscode-remote+ssh-002dremote-002bcn4248.vscode-resource.vscode-cdn.net/data/williamsd7/conda/envs/spatialstuff/lib/python3.12/site-packages/SpaGCN/calculate_adj.py:57) 	nbs=image[max(0,x_pixel[i]-beta_half):min(max_x,x_pixel[i]+beta_half+1),max(0,y_pixel[i]-beta_half):min(max_y,y_pixel[i]+beta_half+1)]
     [58](https://vscode-remote+ssh-002dremote-002bcn4248.vscode-resource.vscode-cdn.net/data/williamsd7/conda/envs/spatialstuff/lib/python3.12/site-packages/SpaGCN/calculate_adj.py:58) 	g.append(np.mean(np.mean(nbs,axis=0),axis=0))
     [59](https://vscode-remote+ssh-002dremote-002bcn4248.vscode-resource.vscode-cdn.net/data/williamsd7/conda/envs/spatialstuff/lib/python3.12/site-packages/SpaGCN/calculate_adj.py:59) c0, c1, c2=[], [], []

TypeError: slice indices must be integers or None or have an __index__ method

Thanks