ML.DIM_REDUCTION.PCA¶
Creates a Principal Component Analysis object.
Syntax¶
ML.DIM_REDUCTION.PCA(n_components, whiten, svd_solver, tol, iterated_power, n_oversamples, power_iteration_normalizer, random_state)
Arguments¶
| Name | Type | Default | Description |
|---|---|---|---|
| n_components | int | None | Number of components to keep. If None, all components are kept. Use 'mle' for automatic dimension selection. |
| whiten | bool | FALSE | If True, scale the components to ensure uncorrelated outputs with unit variance. |
| svd_solver | str | "auto" | Solver to use: 'auto', 'full', 'arpack', or 'randomized'. 'auto' selects the best solver automatically. |
| tol | float | 0.0 | Tolerance for singular values when using 'arpack' solver. |
| iterated_power | str | "auto" | Number of iterations for power method when using 'randomized' solver. |
| n_oversamples | int | 10 | Additional random vectors for 'randomized' solver to ensure proper conditioning. |
| power_iteration_normalizer | str | "auto" | Normalizer for randomized SVD: 'auto', 'QR', 'LU', or 'none'. |
| random_state | int | None | Random seed for 'arpack' or 'randomized' solvers. Use an integer for reproducible results. |
Examples¶
Examples coming soon
Working Excel formula examples for this function are not yet written.