-
Motivated from https://github.com/ViCCo-Group/frrsa/issues/43, if the aim is to generalize the _betas_ to another imageset:
The user submits two (or several) targets and corresponding two (or sever…
-
Possibly change how optionally returned final `betas` are computed to make them more useful for downstream analyses. For that, change `frrsa/frrsa/fitting/fitting/final_model` so that it does a repeat…
-
Both in `g_optim` (https://github.com/rosinality/stylegan2-pytorch/blob/master/train.py#L467) and `d_optim` (https://github.com/rosinality/stylegan2-pytorch/blob/master/train.py#L472) first `beta` is …
-
Is it possible to make a library for the dev version of HA, or at least for the beta versions?
Today we miraculously caught an incompatibility with the new beta. If it hadn't been fixed, many users…
-
Hello,
I am trying to analyse my methylation DNA EPIC v 2.0 arrays with sesame.
I am using windows version of Sesame.
How to solve BiocParallel errors in SeSaMe ? I can't even run the examples su…
-
Hello,
Thank you for the amazing work. I was trying to run the `--save_obj` flag but it doesn't seem to save any output SMPL meshes. I further checked that there has been `mesh_filename` variable t…
-
ser num:108822iten num:94949data ready.
running k-means on cuda:0..
[running kneans]: 0it [00:00，?it/s,center_shift=0.066783,iteration=1, tol=0
[running kneans]: 1it [00:00,10.89it/s,center_shift=0…
-
c65toc64wrapper appears to be not working with a recent environment (development core, ROM beta 982390). This might be better as a ROM bug but I want to accrue notes near the wrapper code.
**Test E…
-
**Describe the bug**
Hi, I am observing that the learning rate suddenly dropped to `model.optim.sched.min_lr` after `model.optim.sched.warmup_steps`. I am using `CosineAnnealing`, where I am expecti…
-
Right now betas have are ahead of the actual release that will be due to a fact that cargo can't handle pre-releases well and assumes that each pre release is compatible with each other, the ways to o…