fix: least_squares implementation #1550
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Outdated suggestions cannot be applied.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.
The Nx.LinAlg.least_squares implementation assumed that for square matrices, it should fall back to Nx.LinAlg.solve.
However, the least squares solution is to minimize norm(Ax - b, 2), which isn't necessarily equal to solving Ax = b.
For example, if A = [[1, 2], [2, 4]] and b = [[3], [5]], Nx.LinAlg.solve will not be able to find a solution as the system doesn't have a valid solution. By going the pinv route, the minimal solution will be found.