genbmm
genbmm copied to clipboard
CUDA kernels for generalized matrix-multiplication in PyTorch
when pip install, report: ``` matmul_cuda_kernel.cu(677): error: expected a ")" 1 error detected in the compilation of "matmul_cuda_kernel.cu". ``` * envirorment visual stduio 2019 cude 11.3 torch 1.11.0+cu113
I install the module by `!pip3 install -qU git+https://github.com/harvardnlp/genbmm`. And then, I run the example code : ```python import genbmm a = torch.rand(10, 3, 4).cuda().requires_grad_(True) b = torch.rand(10, 4, 5).cuda().requires_grad_(True)...
I found log-bmm very useful for linear-chain CRF to save memory and speed up, while in context-free grammars, A->BC requires amounts of GPU memories, which is more serious. So it...
Is it possible to multiply a sparse matrix by a non-sparse matrix? What I'd like to do is multiply a (sparse) transition matrix in an HMM with a (non-sparse) vector...