tensorflow / tensorflow/probability

Scale of ExitBijector

Open
#1,274 0 comments 3 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

https://github.com/tensorflow/probability/blob/v0.12.1/tensorflow_probability/python/bijectors/glow.py#L579

this_scale = scale.Scale(scale_fn(possible_output[..., :c] + 2.))

In calculating the scale of ExitBijector, adding 2 to the log-scale. What is the reason?

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 by reading the referenced ExitBijector code in tensorflow_probability/python/bijectors/glow.py around line 579, then inspect the surrounding Glow implementation and any available documentation for the scale calculation. Done means explaining the purpose of adding 2 to the log-scale, or identifying the documentation location that should clarify it.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, tensorflow
Domain
machine-learning
Issue type
Documentation
Difficulty
3/5
Estimated time
1-2 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.