Skip to content

Shmh40/dev ssl/safe delivrbl tjepa model - #2667

Draft
shmh40 wants to merge 72 commits into
develop-sslfrom
shmh40/dev-ssl/safe-delivrbl-tjepa-model
Draft

Shmh40/dev ssl/safe delivrbl tjepa model#2667
shmh40 wants to merge 72 commits into
develop-sslfrom
shmh40/dev-ssl/safe-delivrbl-tjepa-model

Conversation

@shmh40

@shmh40 shmh40 commented Jul 21, 2026

Copy link
Copy Markdown
Contributor

Description

Issue Number

Is this PR a draft? Mark it as draft.

Checklist before asking for review

  • I have performed a self-review of my code
  • My changes comply with basic sanity checks:
    • I have fixed formatting issues with ./scripts/actions.sh lint
    • I have run unit tests with ./scripts/actions.sh unit-test
    • I have documented my code and I have updated the docstrings.
    • I have added unit tests, if relevant
  • I have tried my changes with data and code:
    • I have run the integration tests with ./scripts/actions.sh integration-test
    • (bigger changes) I have run a full training and I have written in the comment the run_id(s): launch-slurm.py --time 60
    • (bigger changes and experiments) I have shared a hegdedoc in the github issue with all the configurations and runs for this experiments
  • I have informed and aligned with people impacted by my change:
    • for config changes: the MatterMost channels and/or a design doc
    • for changes of dependencies: the MatterMost software development channel

sophie-xhonneux and others added 30 commits May 6, 2026 17:33
Address PR review feedback:
- masking.py: extract self_flow branch body into _generate_self_flow_mask()
- datasets/utils.py: add precompute_cell_ids() (moved from multi_stream_data_sampler)
- masking.py: add Masker.apply_noise_to_data() method that encapsulates
  cell-ID precomputation and noise application (both student and teacher),
  removing the MultiStreamDataSampler's direct dependency on the masking strategy
- multi_stream_data_sampler.py: remove _precompute_cell_ids and _apply_noise_to_data
  module-level functions; delegate to self.masker.apply_noise_to_data()
- multi_stream_data_sampler.py: rename input_data_tgt -> input_data_target

Co-authored-by: Copilot <[email protected]>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

Status: No status

Development

Successfully merging this pull request may close these issues.

5 participants