tensorflow / tensorflow/probability
Functionality for composite 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
I am trying to write up a TensorFlow distribution for the composite lognormal-Pareto model introduced here: https://www.tandfonline.com/doi/abs/10.1080/03461230601110447?journalCode=sact20
Is there a general recipe for writing up new distributions? In particular, I'm having trouble writing the sample() method.
Any help/advice will be appreciated!
Edit: There's a public R package for such distributions as documented here.
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
Start by reviewing the composite lognormal-Pareto model in the linked paper and the referenced R package, then inspect the project's existing distribution implementations and their sample() methods. Done means the requested distribution approach is defined clearly enough to implement, with a documented recipe or a working distribution contribution.
Written by the indexing model from the issue text.
Assessment
- Domain
- machine-learning
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100