category_encoders

category_encoders copied to clipboard

Feature request: min_group_size for other category encoders, e.g. one hot

11

CountEncoder() has a min_group_size parameter that sets a minimum number of obs in a group required in order not for the group to be lumped together with other small groups....

philip-khor

Add Unary/ Thermometer Encoding

1

Add Unary/Thermometer Encoding as an alternative for Ordinal Encoding. According to [Source](https://openreview.net/forum?id=S18Su--CW) it is also more robust to adversarial examples

iki77

[DOC] Gaussian noise regularization: multiplicative or additive?

2

In the documentatin of the gaussia noise regularization it says > adds normal (Gaussian) distribution noise into training data in order to decrease overfitting (testing data are untouched). Sigma gives...

cmougan

Broken inverse_transform for OrdinalEncoder when custom mapping in use!

3

I wanted a start with index 0 instead of 1 hence, custom mapping using enumerate, ``` oe_mapping = [{'col': c, 'mapping': {map_: map_idx for map_idx, map_ in enumerate(df[c].unique())}} for c...

mdalvi

Example of TargetEncoder w/ categorical target

8

Is there an example of using `TargetEncoder` w/ a categorical target variable? The docstring suggests that it should be possible, but I don't see how the code is determining that...

bkj

PolynomialWrapper retains old values from previous calls to fit

1

I think I've found a subtle bug in PolynomialWrapper. The expected behaviour is that if there are n classes in the target column, then PolynomialWrapper will return n-1 new features...

stepthom

feature suggestion: Add Gray encoding

1

## Feature description > Gray Encoding. Our second approach is a variant of binary encoding, namely, Gray encoding [24]. It is a binary encoding system where two successive values differ...

sbrugman

Fix bad WOE scores.

3

I found that the WOEEncoder was not giving the right scores, and it was because of the the agg stats not being correct from pandas. The stats did work if...

slundberg

Implement Target Encoding with Hierarchical Structure Smoothing

13

From section 4 of the [paper](https://kaggle2.blob.core.windows.net/forum-message-attachments/225952/7441/high%20cardinality%20categoricals.pdf) sited in TargetEncoding. > Instead of choosing the prior probability of the target as the _null hypothesis_, it is reasonable to replace it with...

JoshuaC3

Count Encoding

4

Although this is very simple to implement in Pandas or similar, it would be very nice to have this in here as a scikit compatible transformer. It is a good...

JoshuaC3