-
Hi Matthias! Terrific package, I've been using staRdom to completely eliminate my lab's need for expensive standalone software.
I've been trying to run some split-half analyses, but the function j…
-
Very interesting work! A quick question, can we understand this method as a generalization of factorization machine (https://cseweb.ucsd.edu/classes/fa17/cse291-b/reading/Rendle2010FM.pdf)? it appears…
-
Hey everyone. Just wanted to add a feature idea here.
When using PARAFAC to process EEM data, it is very common to fix two modes (namely the modes for emission and excitation) to those values previ…
-
Will require that this is done being implemented in Tensorly (with masking). This will allow us to incorporate glycosylation.
https://github.com/tensorly/tensorly/pull/151
-
I have a sparse COO array that I'm attempting to decompose in a sparse manner.
Here is my input array.
``````
I don't appear to have a "sparse" option for a backend, when choosing backend…
-
Just a quick bug report concerning various incoherent uses of 10e-xx and 1e-xx. For instance,
tl.decomposition.test.test_candecomp_parafac.py contains lines such as
` tol_norm_2 = 10e-2 `
but insi…
-
I'd like to put in a request (similar to #5) for support for `PARAFAC2`, to handle irregular tensors with different sizes in one mode and shifts in dynamics.
I assume this has been on the to-do lis…
-
Hello,
#### Describe the bug
When running PARAFAC with a PyTorch backend on a tensor, the reconstruction error increases on every iteration, but only with certain tensors and only with missing v…
-
Hi, like milanlanlan in [Handling missing data in decomposition #4](https://github.com/tensorly/tensorly/issues/4#issuecomment-557899752) I am trying to decompose a tensor using the CP decomposition w…
-
Hi, One of the use case of matrix and tensor factorization is in movie recommendation where the matrix/tensor are sparse. I tried Tensorly with missing data and it fails.
I was wondering if handlin…