-
#### Describe the bug
When applying the code to my own project, I noticed I was not getting the expected output which is described in the documentation as a tuple containing weights and factors. It…
-
It would be good to add enhanced Line Search [1], at least to CP, as proposed in #29.
[1] M. Rajih, et. al., "Enhanced Line Search: A Novel Method to Accelerate PARAFAC", SIAM J. Matrix Anal. Appl.…
-
Hello Filip,
I wanted to open a different issue for this question.
We are very happy with candia's capabilities so far and would probably be testing it soon with a bigger cohort of samples.
W…
-
#### Describe the bug
Error when using init='random' with pytorch backend and RTX 3090 GPU
This occurred after upgrading from torch 1.6.0 to 1.7.0
The below code works when switching to init='svd…
-
#### Describe the bug
When I try to use mask parameter in parafac function for missing values, it doesn't work.
#### Steps or Code to Reproduce
`factors = parafac(tensor=mask_img, init='rand…
-
Hello,
We are very excited about CANDIA's capabilities and are interested in testing it for our data analysis pipelines. I wanted to share my current experience with the installation and execution …
-
Hi,
In an end-to-end deep learning framework with **SGD** optimizer, I'm calling the function "**tucker**" on a 4D tensor (of size **200, 128, 4, 4**) with pytorch in the backend. The rank parameter…
-
Hi Matthias,
Thank you for your package first of all!
I have an issue with parafac_conv. When I try to run it with the parameter output="all" it returns this:
Error in parafac_conv(eem_arra…
-
I decided to try the sparse CP from contrib.sparse.decomposition but got ModuleNotFound error.
I am simply doing this -
`import tensorly as tl`
`from tensorly.contrib.sparse.decomposition import…
-
Hello,
I noticed that there is an _orthogonalize_ parameter in parafac.
`parafac(tensor, rank, n_iter_max=100, init='svd', svd='numpy_svd', normalize_factors=False, **orthogonalise=False**, tol=1e…