Ioannis Siglidis

Some guidelines for better gpu perfomance.

2

Hi, I am trying to train on a custom dataset using your algorithm. I try to increase batch size up to the point that my script doesn't break. I am...

Hi, the only example I find in your code of class conditioning seems to be [cin256](https://github.com/CompVis/latent-diffusion/blob/main/models/ldm/cin256/config.yaml). However there, conditioning is not performed through concatenation but rather through cross-attention. I couldn't...

Hi, I think I may found a small bug in the visualization part of ddpm. Except the case where the mask changes in the sample_log function (which from what I've...

How to sample high-resolution images?

3

Hi, this paper claims to be able to produce high-resolution images, yet there is no configuration in the code that can learn images bigger that 256x256. Increasing the resolution to...

`sample_fast` images are incoherent with logged `no_pix` images during train.

2

Hi, While training the Net2Net transformer model, after I have trained the VQ-GAN, I see interesting images being generated on the logs, with `no_pix` in their name, which probably means...

Instructions for training a performance_rnn

1

Hi, I am trying to run a performance rnn on a jazz dataset, but training doesn't seem at all to generalize. I am using the configuration `performance_with_dynamics`. Basically the maximum...

Add a rows mode or a functionality to transpose the table.

2

Hi and thank a lot for your awesome package. I discover, that for some cases of outputs it would really useful if you could have on the same col different...

st.map() : display image on hover over scattered points

1

Extremely simple use case: I want to have a map that to which I can hover over points and see the image associated to each point. This is easy to...

Weirdly enough there seems to be no tensorboard logging and validation. Is this the case or I didn't run something correctly?

First of all thanks a lot for this wavenet code, well written and handed to the community. My problem is that I cannot understand, if my model is learning and...