-
The PixelCNN learn to model the prior q(z) in the paper and the code. For any given classes/labels, PixelCNN should model their prior q(z), as shown in the code https://github.com/ritheshkumar95/pytor…
-
Hi. When I'm using ESM3, I find the reconstruction performance of the provided structure vqvae is poor (rmsd about 10). Here is my code. Is there any misunderstanding or misuse of the method?
```
f…
-
Hi,
Thank you for your work, I have no access to the pretrained vqvae model:)
Is any possible to get access?
-
# Neural Discrete Representation Learning #
- Author: Aaron van den Oord, Oriol Vinyals, Koray Kavukcuoglu
- Origin: https://arxiv.org/abs/1711.00937
- Related:
- https://avdnoord.github.io/…
-
Great job! I’d like to ask a question about training loss. In the process of training diffusion model, there are two losses: $L_{LD}$ and $L_{AIR}$ (corresponding to Equation 8 and Equation 10 in the …
ddz16 updated
2 months ago
-
Hi! I'm failing to understand the function of PixelSnail. **Is it to generate a latent space similar to a GAN?**
I trained VQVAE correctly (until the samples were good enought):
`python train_vq…
-
model = '5b_lyrics' # or '5b' or '1b_lyrics'
hps = Hyperparams()
hps.sr = 44100
hps.n_samples = 3 if model in ('5b', '5b_lyrics') else 8
# Specifies the directory to save the sample in.
# We set …
-
### What happened?
ImportError: cannot import name 'VectorQuantizer2' from 'taming.modules.vqvae.quantize'
File "/content/stable-diffusion/optimizedSD/ddpm.py", line 14, in
from ldm.models.a…
-
Hello ESM team,
Thank you for releasing the ESM3 code! I have a question regarding the training protocol you have described for the VQ-VAE model.
In the paper, you describe a two-stage training …
-
Hi, thanks for sharing the great work.👍
I am trying to reuse the motion tokenizer (VQVAE) in MotionGPT so I wonder.
1. Is there a way to save the VQVAE's weight as a checkpoint file separately fro…