click below
click below
Normal Size Small Size show me how
Prepare data for ML
Prepare data for Machine learning algorithms
| Question | Answer |
|---|---|
| Reasons to prepare data for ML algos? | * Reproduce transformations on any dataset * Build a library of transformation functions that you can reuse * Use these functions in your live system to transform new data before feeding it to your algorithms * Easily try various transform and see whic |
| Data cleaning options for attributes | 1. Get rid of the corresponding rows containing the missing attribute 2.Get rid of the whole attribute 3.Set the values of some value (zero,the mean, the median,etc.) |
| Scikit-learn Design: main design principles | * Consistency * Inspection *Non proliferation of classes * Composition * Sensible defaults |
| Sciket-learn: Consistency principle | All objects share a consistent and simple interface |