Comments (2)
Hi @zen-d,
Thanks for reaching out. I've made the same experiment using the parameters with the same command, and it yield approximately 7 FID (as you mentioned). However, it's worth noting that FID measurements can exhibit some degree of variability, and I did put the best values I had during the evaluation, (i.e., 6.80 FID).
Since I released the code, I would advise changing the hyper-parameters for sampling to achieve a more stable FID. Would you mind trying it on your machine to check the results? I achieved an FID score of 6.85 with cfg_w set to 2.7, r_temp at 7, and 10 steps.
Size of model autoencoder: 72.142M
Acquired codebook size: 1024, f_factor: 16
load ckpt from: /home/victor/workspace/github_release/MaskGITpytorch/pretrained_maskgit/MaskGIT/MaskGIT_ImageNet_256.pth
Size of model vit: 174.161M
Evaluation with hyper-parameter -->
scheduler: arccos, number of step: 10, softmax temperature: 1.0, cfg weight: 2.7, gumbel temperature: 7.0
{'Eval/fid_conditional': 6.849918568671455, 'Eval/inception_score_conditional': 215.3447265625, 'Eval/precision_conditional': 0.81296, 'Eval/recall_conditional': 0.53042, 'Eval/density_conditional': 1.2149066666666668, 'Eval/coverage_conditional': 0.8412800000000001}
from maskgit-pytorch.
@llvictorll Yes, the carefully selected hyperparameters results in a better performance (close to reported).
from maskgit-pytorch.
Related Issues (11)
- Regarding training a mask model with my own data, could you please provide guidance on the steps involved HOT 1
- train vqgan HOT 3
- Sampling with CFG = 0 HOT 2
- About the training intermediate result. HOT 1
- Warm-up of CFG weight HOT 2
- Target tokens for loss computation HOT 2
- Unneccesary Dropout layer in FeedForward network HOT 2
- Has anyone successfully run the code HOT 2
- How can I use my own dataset to train maskgit? HOT 2
- questions about two stage training HOT 8
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 maskgit-pytorch.