-
**Motivation**
Fused multiply–add (FMA) is a floating-point operation performed in one step, with a single rounding. FMA can speed up and improve the accuracy of many computations: dot product, mat…
-
#### Description:
The FFTW library can be used seamlessly with Eigen and their FFT implementation which makes everything simple to use with Stan. Using FFTW has proven to yield huge speedups in our m…
-
### 🚀 The feature
I propose to open the discussion and collect in this issue some discrepancies or duplicate functionalities I found between detectron2 and torchvision.
detecton2 / mmdet / korni…
-
hi,I'm from sd.cpp.I am very interested in the Winograd convolution algorithm you mentioned, and I'd like to know how its progress is going. I wonder why it's no longer on the sd.cpp to-do list.
-
Since Theano now has several implementations for conv2d and conv3d, I wonder if there is any support for adding a dedicated 'conv1d' op.
I have been using Theano for applications where 1D convolution…
-
Currently, `SpatialConvolutionMM` is quite fast on CPU, but its memory requirements are too high.
As it parallellizes the computations over batch examples, it requires a huge buffer (dependent on the…
-
Hello!
I am interested in using Julia/imfilter for a deconvolution problem in particle physics. In short, our experiment (http://next.ific.uv.es/next/) takes "electronic movies" of electrons propag…
-
Has anybody tried to bring LZ to the web?
There is this new WebAssembly thing. We could take some C library that implements all the needed ML operations (convolutions, etc.), feed it the weights, u…
klueq updated
4 years ago
-
I've thinking about the possibility of extending this package to be able to perform Vector/Tensor calculus operations.
A suitable discretization of a vector field is an `Array{SVector{N, T},N}`.
…
-
https://arxiv.org/abs/1801.01671