tensorflow / tensorflow/probability
Independent Layer with Normal and Bernoulli distributions
Nobody has claimed this yet.
- Dominant language
- Jupyter Notebook
- Stars
- 4.4k
- Forks
- 1.1k
- PR merge metrics
- No merged PRs in 30d
Description
Is it possible get an Independent Layer with Normal and Bernoulli distributions as the output layer of a variational autoencoder?
I am trying to detect anomalies in time series with a variational autoencoder. But I have binary signals as well as continuous signals. I would like to get one single sample from the model just like I would get from an IndependentNormal, but use Bernoulli distributions for the binary signals.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
No files, tests, or entry points are named. Start by locating the existing IndependentNormal layer and related variational autoencoder distribution layers, then determine how mixed continuous and binary outputs should be represented. Done would require a clear supported approach for producing one sample with Normal and Bernoulli components, along with validation of that behavior.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- tensorflow
- Domain
- machine-learning
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 20/100