feature_engine

feature_engine copied to clipboard

RecursiveFeatureElimination Error using Ccross-validation Splitter

4

**Describe the bug** When I try to use a cross-validation splitter instead of an integer value in `RecursiveFeatureElimination` I face an error that says `IndexError: list index out of range.`...

behzad89

Features/distances

15

Just a first sketch. Let me know what do you think :-)

VascoSch92

`pseudolog` transformer

9

Request for an inverse-hyperbolic-sine, a.k.a. asinh or pseudolog, transformer. `x->arcsinh(x/2)` behaves like `ln(x)` for large values of `x`, but behaves like `x->x` for small values of `x`; this behavior is...

ParadaCarleton

Add missing_only init param to all imputers. missing_only is used in combination w/ the variables param.

10

Closes #388. When `missing_only=True` and `variables=None`, all numerical variables that do **not** contain missing values will be omitted from `self.variables_`. This functionality does not apply to categorical variables.

Morgan-Sell

add test to ensure estimators raise errors when X has duplicated column names

5

Per comment in PR: https://github.com/feature-engine/feature_engine/pull/686#pullrequestreview-1544746361 After the PR in the link above gets merged, there will be a general check for duplicated columns for which a test was added in...

david-cortes

future warnings

3

List of future pandas warnings ``` c:\users\sole\documents\repositories\feature_engine\feature_engine\creation\math_features.py:212: FutureWarning: The provided callable is currently using Series.sum. In a future version of pandas, the provided callable will be used directly. To keep...

solegalli

Cyclical features when variables are negative - does it make sense?

6

Not sure we can make a variable with negative values cyclical with the sine and cosine. We need to investigate more and if it does not make sense, we need...

solegalli

Geo distance transformer

4

Hei, For a project, I created a transformer that given the coordinates of two points on the earth calculates the distance between them. It would be interesting to have in...

VascoSch92

- all discretizers apart from arbitrary - all imputers - boxcox, log, power reciprocal and yeojohnson - wrappers Either replace by a dataset from sklearn, or see if we are...

solegalli

not sure it is possible, and it is certainly a major undertaking but: Is it possible to use something like partial_fit for feature engineering? So that we can allow feature...

solegalli