-
Implement Fast Fourier Transformations for convolutions
-
## 🚀 Feature
An FFT backend for `kornia.filters.filter2d`.
## Motivation
`kornia.filters.filter2d` is very slow for large kernels as it currently only performs convolutions in the spatial domai…
-
What are your iteration counts on the CelebA-HQ dataset? Why does my model perform well on the training set but poorly on the test set? Could you please provide your training strategy?
-
I have implemented an extremely dumb version of convolution,
```
u * v = 1/volume sum_a u[a] v[a-r]
```
by constructing a rank-3 tensor that I call a `convolver`, such that
```
u *…
-
Hello,
Thank you for the effort you put into making this work, however I am very confused. When I want to apply this "FFTCONV2D" layer to a network (Resnet for example), in GPU I always get the err…
-
It would be very cool to be able to perform convolutions on a selected area with user-specified kernels. Properly implemented, this could be really useful to easily and seamlessly smooth out landscape…
-
Hello, how should I write a prompt to create the Seamless Repeat Pattern?
I tried the Flower Rose Seamless Pattern prompt and many other prompts but the perfect pattern does not come out. Irrelevant …
-
Does p3dfft support the computation of convolutions
-
It would be great to propose several sets of wavelets. Gabor are real in fourier, thus adding Haar wavelets (despite being simple and having a different structure) could make a lot of sens w.r.t. the …
-
By adding this test:
```python
def test_wavelet_adjoint(wave_impl, shape_setup, floating_dtype):
"""Verify that the adjoint of the wavelet operator is correct.
This is done by checki…