issues
search
harvardnlp
/
genbmm
CUDA kernels for generalized matrix-multiplication in PyTorch
79
stars
13
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
matmul_cuda_kernel.cu(677): error: expected a ")"
#14
www008
opened
2 years ago
1
NameError: name '_genbmm' is not defined
#13
HenryBao91
opened
2 years ago
1
RuntimeError: CUDA error: an illegal memory access was encountered
#12
haozheji
closed
2 years ago
4
Run into errors when building wheel for genbmm
#11
haozheji
closed
2 years ago
4
Add example tests
#10
JohnReid
closed
2 years ago
1
Fix too many values to unpack.
#9
JohnReid
closed
2 years ago
0
Fix backward pass bugs, add fp16 support, add product-max
#8
keyonvafa
closed
3 years ago
0
About the performance
#7
speedcell4
closed
3 years ago
6
[feature request] support log-bmm to context-free grammars
#6
sustcsonglin
opened
4 years ago
22
Write higher-order semirings
#5
srush
closed
4 years ago
0
Non-sparse x sparse
#4
lorenlugosch
opened
4 years ago
2
Support for `half`
#3
codekansas
closed
3 years ago
2
Unable to compile matmul_cuda_kernel.cu
#2
MaximumEntropy
closed
4 years ago
1
Banded sparse matrix.
#1
srush
closed
4 years ago
0