Issue with automatic mesh generation for selected segmentations in Neuroglancer

[yuxuandexter]

I load h5 file from the local directory to the neuroglancer, the neuroglancer could load all instance segmentations but it failed to automatically generate a mesh.

[yuxuandexter]

After further debugging, I found out that the neuroglancer can not support 32 bit automatic mesh generation in the default. Is there any way to solve this problem?

[jbms]

You need to change the layer type to segmentation --- click on where it says img next to the layer name and select segmentation.

[yuxuandexter]

Thanks for your reply. I figured out my minor mistake. However, I still want to know whether neuroglancer support 32 bit automatic mesh generation.

Based on this image, I double click the selected segmentation, which fails to generate a 3D mesh. My dataset is 32 bit for segmentation. Is there any available way to solve this problem?

[jbms]

int32 isn't supported but uint32 is.

[yuxuandexter]

Thank you so much. One last question. Do you know where is the neuroglancer python API documentation? If I want to add proofreading function, what type of resources will be helpful for me to do it!