use the target random number for also source random number

[kctezcan]

Description

Problem: the random number used to make the noise signal for the diffusion forward process is coming from the source random draw. the random number for calculating the loss weight for diffusion is coming from the target random draw.

Fix: take the arget random number also for the source.

Remark: for this to work, the target_input in the config has to be set:

  target_input: {
    "forecasting" : {
      ...
      masking_strategy_config: {
        diffusion_rn: True,
        noise_distribution: *noise_distribution,
        sigma_min: *sigma_min,
        sigma_max: *sigma_max
        },
      ...

Issue Number

No issue

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

[clessig]

@kctezcan : if this problem also exists on develop-ssl-diffusion, could we also fix it there?

[kctezcan]

@kctezcan : if this problem also exists on develop-ssl-diffusion, could we also fix it there?

do you mean in develop-ssl-diffusion-v1?

[clessig]

@kctezcan : if this problem also exists on develop-ssl-diffusion, could we also fix it there?

do you mean in develop-ssl-diffusion-v1?

Yes

[Jubeku]

Do we have a problem here for forcing streams which don't have a target mask and hence no random number? Maybe in this case my PR 2737 is complementary because a target mask is built even if target channels are empty.