tensorflow / tensorflow/probability
Hidden semi markov model using tfd.HiddenMarkovModel
Nobody has claimed this yet.
- Dominant language
- Jupyter Notebook
- Stars
- 4.4k
- Forks
- 1.1k
- PR merge metrics
- No merged PRs in 30d
Description
Dear maintainers,
I was wondering if there is any way of extending tfd.HiddenMarkovModel to a hidden semi-Markov setting, where emissions depend not only on the last seen hidden state but also on the time elapsed since the last state switch (see figure below).
Thanks a lot for your help!
Lucas
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 reading the tfd.HiddenMarkovModel API and the issue’s hidden semi-Markov diagram to determine whether duration-dependent emissions fit the existing model. Define the required extension and its expected behavior before identifying implementation files or tests; done would mean a supported hidden semi-Markov model with documented duration-dependent emissions and validation.
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