Uniform Noise Adder using LoopBuilder #115
Merged
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This work is a quick cleanup of a stale branch adding a uniform noise Op, as well as a simple regression test. Some points on design decisions:
RandomAccessibleIntervals and makes use of theLoopBuilderconstruct. For simplicity, computation is performed sequentially - I didn't see a clear pathway to run this in a multithreaded way, because we'd have to (a) use a thread-safe RNG object, which would be slower, and I didn't bother figuring out whether it would be worth it, or (b) create aMersenneTwisterFastper chunk, which is tricky because as far as I can tellChunks give you no discernible state that we could use to seed eachMersenneTwisterFastdifferently.RealTypehow out-of-bounds values are handled. There are also situations where we may have lossy behavior - for example, if the range provided is very large - but I think that's a hard problem to solve anyways.I honestly think this is a pretty low-priority solve - but at the very least, we've moved broken code from a stale branch to a PR that could be closed without losing anything important.