Tristan Bepler
Tristan Bepler
Of course, I'm glad to hear the code release is helpful. To generate images from the model, you need to select some point in latent space ($z$) and then use...
I was referring to the unstructured latent variables with _z_. These are set to have standard normal prior by the training procedure. If you want to perform inference on _z_...
Can you share the exact command you ran and your python environment (e.g., `conda list`)?
I believe this is already fixed on main, but we haven't pushed that out in a new version on conda/pip. Have you tried installing from source?
The pretrained 2d checkpoint model was trained on a mix of a few labeled datasets from EMPIAR and a much larger set of simulated single particle micrographs, which is not...
It's been a long time, but I'll see if I can dig up the complete list. It was not a lot of EMPIAR datasets (~5) and definitely included EMPIAR-10025 and...
I removed the `seed` argument because of multiprocessing. If the random number generator is initialized within each process, you will not get the same result when processing a list of...
@swan343 my main concern is actually about consistency between runs with different numbers of workers or no workers. With the same number of workers, the results should be the same,...
Ah, I see. You want to reset the random number generator to the same seed for every image. This means every image will start from the same random initialization. Is...