scDeepCluster_pytorch icon indicating copy to clipboard operation
scDeepCluster_pytorch copied to clipboard

Published 20 hours ago • verified publisher icon ttgump

RuntimeError during clustering step

Open hellorp1990 opened this issue 1 year ago • 2 comments

Hi,

Thanks for the awesome tool. I ran the training model step successfully. But I encountered an issue during the "Clustering stage".

RuntimeError: "normal_kernel_cpu" not implemented for 'Long' Please check the screenshot. Capture

Please let me know how can I fix it the error.

hellorp1990 avatar Nov 17 '23 14:11 hellorp1990

Could you let me know the version of PyTorch that you are using?

ttgump avatar Nov 27 '23 15:11 ttgump

I checked the message you posted. It seems the error comes from the line 69 in scDeepCluster.py that uses "torch.randn_like" function. "torch.randn_like" function will sample random values from the Gaussian distribution, so it should be float values. But in your case, your input data x seems like to be a Long data type. Input x should be float32 or float64 values. Could you check your input data?

ttgump avatar Nov 27 '23 16:11 ttgump