probml / probml/dynamax

pass inputs into the LDS model

Open
#329 3 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
1k
Forks
114
Avg merge
19h 14m
Merged PRs (30d)
1

Description

Hello,

I have a very basic question: how to pass N X T X D inputs ("X") into the LDS model (N trials, T time steps and D dimensional inputs)?

In the linear_gaussian_ssm model.py file, the inputs is Optional[Float[Array, "ntime input_dim"]], so there's no dimension for trials (N)?

I tried to do things as in the Kalman filter/ smoother example. But the problem is that I also need to include d latent trajectoreis into the model (i.e. the state dimension should be D + d, if I encode the covariates into the emission matrix).

Not sure how to do it correctly...

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

Start with the linear_gaussian_ssm model.py file and the Kalman filter/smoother example, checking how inputs are shaped and whether trial batching is supported. Trace how the model handles latent trajectories and define the expected N × T × D interface; done means the supported behavior is implemented or clearly documented with a corresponding example or test.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
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.