Remove unnecessary restriction on number of samples in IncrementalPCA

[ThomasGesseyJonesPX]

Currently when calling IncrementalPCA.partial_fit() the number of samples in X always has to be greater or equal to the number of PCA components due to this hardcoded check

scikit-learn/sklearn/decomposition/_incremental_pca.py

Line 303 in 6e90391

elif not self.n_components <= n_samples:

However, this restriction actually only needs to apply to the first call to partial_fit(), which performs the initial SVD decomposition. Once this first call has established the initial $U$ and $\Sigma$, none of the algorithms steps (see original paper https://www.cs.toronto.edu/~dross/ivt/RossLimLinYang_ijcv.pdf) require any size restrictions on the number of samples in the subsequent batches passed to partial_fit. The current error message hence unnecessary restricts the usage of IncrementalPCA, in particular in online applications where data might arrive in batches of different sizes.

This PR, updates the aforementioned check to only apply to the first call of partial_fit and adds a non-regression test for this issue.

[github-actions]

✔️ Linting Passed

All linting checks passed. Your pull request is in excellent shape! ☀️

_{Generated for commit: c752837. Link to the linter CI: here}

[ogrisel]

Thanks for the PR. Some minor feedback below, but otherwise LGTM.

We probably need better tests (e.g. on non-spherical data) for this estimator, and we could also clarify what are the expectations in terms of accuracy between the full-batch and incremental version of PCA, but this is beyond the scope of this PR.

[ThomasGesseyJonesPX]

Thanks for the feedback! I have implemented all of your suggestions