A fast, ergonomic and portable tensor library in Nim with a deep learning focus for CPU, GPU and embedded devices via OpenMP, Cuda and OpenCL backends
I just tried to create an identity matrix and the program keeps crashing: ``` import arraymancer var b = eye[float](3, 3) echo b ``` or ``` import arraymancer var b = identity[float](3) echo b ``` The output: ``` /home/~/tester.nim(12, 24) template/generic instantiation of `identity` from here /home/~/Arraymancer/src/arraymancer/linear_algebra/special_matrices.nim(238, 16) template/generic instantiation of `diag` from here /home/~/Arraymancer/src/arraymancer/linear_algebra/special_matrices.nim(226, 9) template/generic instantiation of `set_diagonal` from here /home/~/Arraymancer/src/arraymancer/linear_algebra/special_matrices.nim(172, 26) Error: type mismatch Expression: &"Diagonal index ({k=}) exceeds input matrix height ({a.shape[0]})" [1] "Diagonal index ({k=}) exceeds input matrix height ({a.shape[0]})": string Expected one of (first mismatch at [position]): [1] func `&`(a, b: DynamicStackArray): DynamicStackArray [1] func `&`[T](a: DynamicStackArray[T]; value: T): DynamicStackArray[T] [1] proc `&`(x, y: char): string [1] proc `&`(x: char; y: string): string [1] proc `&`[T](x, y: sink seq[T]): seq[T] [1] proc `&`[T](x: sink seq[T]; y: sink T): seq[T] [2] proc `&`(x, y: string): string missing parameter: y [2] proc `&`(x: string; y: char): string missing parameter: y [2] proc `&`[T](x: sink T; y: sink seq[T]): seq[T] missing parameter: y ```
This issue appears to be discussing a feature request or bug report related to the repository. Based on the content, it seems to be still under discussion. The issue was opened by general-rishkin and has received 0 comments.