enalisnick / stick-breaking_dgms Goto Github PK
View Code? Open in Web Editor NEWDeep Generative Models with Stick-Breaking Priors
Home Page: http://arxiv.org/abs/1605.06197
License: MIT License
Deep Generative Models with Stick-Breaking Priors
Home Page: http://arxiv.org/abs/1605.06197
License: MIT License
how can i generate image from this model,
how could i do a prediction?
Hi!
I am trying to replicate the results of a standard Stick Breaking VAE (unsupervised) in Tensorflow. One issue I found was even after using the same prior as in your Theano code, the KL divergence came out negative and then the optimizer rapidly reduced it to large to negative values and finally NaN. This corresponded to very small values of alpha and beta (around 0.001s or lower).
I solved that problem by clipping alpha and beta to between 0.25 and 15. However, am not getting good sampled images on using 50 dimensional latent space. I am noticing that though the negative log likelihood i reducing, the KL divergence increases from about 1 to 4. Did you encounter similar issues and if so, what did you do to solve them?
Hi!
I am a bit confused after reading the following line in code.
Does this mean that the a
and b
for Kumaraswamy distribution are different for each dimension? Aren't the \alpha and \beta for the Beta distribution in SBP same for all v
s?
Also, you used latent_size-1
at many places. Why -1
?
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.