theislab / theislab/interscale

Masking in Transformer

Open
#43 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

enhancement
Dominant language
Python
Stars
12
Forks
1
Avg merge
12h 51m
Merged PRs (30d)
1

Description

Description of feature
  • #18
  • Long-range interaction masking: In mask from torch.nn.transformerencoder set attn_mask to -inf. Currently, not working because together with src_padding_mask it returns NaNs in the global embedding

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 locating the Transformer encoder call and its mask handling, then inspect how attn_mask and src_padding_mask are combined. Reproduce the global-embedding NaNs with both masks enabled and verify that long-range interactions are masked without producing NaNs.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Feature
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
45/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.