Why choose CSR or CSC before passing sparse features to a scaler?
Classic MLLevel: IntermediateSparse storage copies
Why choose CSR or CSC before passing sparse features to a scaler?
Hints
- Consider representation conversion.
- Peak memory includes temporary copies.
Answer
These formats are directly supported; other sparse formats may be converted to CSR. Choosing upstream avoids unexpected conversion copies and peak memory.
Source and licence: scikit-learn 1.7.2 — preprocessing · BSD-3-Clause · answer: Quantoros editorial