Pytorch's linear functions mainly encapsulate Blas and Lapack, with similar usage and interfaces.
The commonly used linear functions are as follows:
| function (math.) | functionality |
| trace | Sum of diagonal elements (trace of matrix) |
| diag | diagonal element |
| triu/tril | Upper/lower triangles of the matrix with specified offsets |
| mm/bmm | Matrix multiplication, matrix multiplication of batch |
| t | reprovision |
| dot/cross | Inner product/outer product |
| inverse | Finding the inverse matrix |
| svd | singular value decomposition (math.) |
Note: The transpose of the matrix will make the storage space discontinuous, you need to call its .contiguous method to turn it continuous.
Example:
import torch as t b=() b.is_contiguous() exports：False b=() b.is_contiguous() exports：True
Above this pytorch commonly used linear functions in detail is all that I have shared with you, I hope to give you a reference, and I hope you will support me more.