Comments (2)
Hi, thank you for testing the code and reporting your result. I think the divergence between training loss and FID was due to overfitting to the dataset and it was expected after FID hit its optimal. Similar observations have been made by many others and myself (e.g., openai/improved-diffusion#91 (comment))
from ddpm-torch.
Thank you for the explanation again :)
from ddpm-torch.
Related Issues (16)
- Vector-conditioning using classifier-free guidance HOT 4
- DDIM sampler training parameters HOT 2
- Question about FID calculation HOT 2
- Bug in ddim.py HOT 1
- FID for CIFAR10 checkpoint HOT 2
- what changes would be needed if we want to use our own custom greyscale(channel=1) dataset having size of 256*256. HOT 1
- some checkpoint HOT 2
- Bug in target noise HOT 1
- CIFAR 10 config HOT 1
- Comment in code is wrong. Upsample supports align_corners=True HOT 2
- Meet bugs when adopting the --eval choice for training. HOT 2
- train my own image datasets HOT 1
- pretrain HOT 1
- Have you tested the FID for CelebaHQ? HOT 2
- Minor bug in Gaussian Diffusion implementation (q_mean_var)? HOT 2
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from ddpm-torch.