You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hello, when we were running the create_heatmaps.py code, we got this error:
Traceback (most recent call last):
File "/home/galen/CLAM/create_heatmaps.py", line 369, in
compute_from_patches(wsi_object=wsi_object,
File "/home/galen/CLAM/vis_utils/heatmap_utils.py", line 82, in compute_from_patches
A[score_idx] = score2percentile(A[score_idx], ref_scores)
ValueError: could not broadcast input array from shape (15882,) into shape (1,)
We noticed that the shape of the ref_scores is [15882, 1], so we resolved this error by flattening the ref_scores. We also noticed that there are several closed issues mentioning this problem with the solution similar to ours. Since these issues are closed, we doubted ourselves whether we made a previous mistake in the pipeline that caused this error. Or are the codes not fixed?
The text was updated successfully, but these errors were encountered:
Hi I am having the same problem, I do not know why we need to reshape the ref_scores and I am not sure whether it is a serious problem to lean on. Some clarification would be very nice ! :)
Hello, when we were running the create_heatmaps.py code, we got this error:
Traceback (most recent call last):
File "/home/galen/CLAM/create_heatmaps.py", line 369, in
compute_from_patches(wsi_object=wsi_object,
File "/home/galen/CLAM/vis_utils/heatmap_utils.py", line 82, in compute_from_patches
A[score_idx] = score2percentile(A[score_idx], ref_scores)
ValueError: could not broadcast input array from shape (15882,) into shape (1,)
We noticed that the shape of the ref_scores is [15882, 1], so we resolved this error by flattening the ref_scores. We also noticed that there are several closed issues mentioning this problem with the solution similar to ours. Since these issues are closed, we doubted ourselves whether we made a previous mistake in the pipeline that caused this error. Or are the codes not fixed?
The text was updated successfully, but these errors were encountered: