tensorflow / tensorflow/probability

HiddenMarkovModel Binomial Extension

Open
#658 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Jupyter Notebook
Stars
4.4k
Forks
1.1k
PR merge metrics
No merged PRs in 30d

Description

Hi,

I'm currently trying to implement a HMM model in which the observation distribution is binomial.
I see in the current tfp.Distributions.HMM the

observation distribution is A tfp.distributions.Distribution-like instance. The rightmost batch dimension indexes the distribution of each observation conditioned on the corresponding hidden state.

If the observation distribution is not only conditioned on the corresponding hidden state but also the current time-step, what would be the recommended way of implementing this under the tensorflow-probability framework?

Thanks for any suggestions.

BR,
Rlyu

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

The question concerns tfp.distributions.HMM and its observation distribution. Start by reviewing the HMM observation-distribution API and how it handles hidden-state and time-step conditioning. Done means documenting or establishing a supported implementation approach for binomial, time-dependent observations.

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
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.