Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

Doesn't seem to generalize well

See original GitHub issue

I tried to test the performance of DDRM by adding a photo of myself to the ood_celeba directory and running

python main.py --ni --config celeba_hq.yml --doc celeba --timesteps 20 --eta 0.85 --etaB 1 --deg sr4 --sigma_0 0.05 -i celeba_hq_sr4_sigma_0.05

While the examples from the original CelebA set are reconstructed very accurately, results on this additional photo are less than optimal, let’s say 😃 Now I’m not a celebrity, but I was wondering whether you have an idea of what is happening here me-face 0_-1 .

Issue Analytics

State:
Created a year ago
Reactions:1
Comments:5 (1 by maintainers)

Top GitHub Comments

1reaction

jiamingscommented, Jul 1, 2022

There is a FFHQ checkpoint from https://github.com/jychoi118/ilvr_adm/ that you might be interested in checking out as well. It is not too hard to add this into our code, as their model is also based on the OpenAI’s ADM model. Afaik, the CelebA model provided by the DDPM authors has some color shift issues if you try to use it for unconditional generation.

1reaction

andreaferretticommented, Jun 29, 2022

Same issue with Scarlett Johansson

2_-1

Top Results From Across the Web

What should I do when my neural network doesn't generalize ...

The most important part is understanding why your network doesn't generalize well. High-capacity Machine Learning models have the ability to ...

What should I do when my neural network doesn't ... - Quora

1. You don't have enough data · 2. Your dataset has one of numerous issues · 3. Network too deep · 4. Network...

Lecture 9: Generalization

If an algorithm works well on the training set but fails to generalize, we say it is overfitting. Improving generalization (or preventing overfitting)...

Training doesn't generalize to predict validation and test set ...

I have built a 1D CNN regression model using features from the embedding layer by keras , to predict a score of some...

Generalization, Regularization, Overfitting, Bias and Variance ...

If validation loss decreases as well, the learned patterns seem to generalize. Bias. Bias is defined as the average squared difference between ...