feature-selector

feature-selector copied to clipboard

In method _identify_collinear_ I discovered, that you do not respect pvalue of Pearson coefficient. That is, one can remove features, which correlation have nor statistical importance \ It can be...

EugeniaKoKo

I found that the auc obtained in the 10 trainings are different, is it possible to remove the auc which is bad, and accumulate the remaining feature_importances_ Thank you

wd1969

Removing my label column during removal?

2

I'm at a loss as to what's happening here. I'm downloading historical stock data with Pandas Datareader, and after some small manipulations (ie. re-arranging the dataframe, adding moving averages, etc.),...

windowshopr

Perform transform on test set

2

Hi, This project looks really cool and important. A good methodology of any data transformation and especially feature extraction is to do it on the train set and then transform...

shaygeller

I have query regarding feature selector's zero importance and low importance object. Is the metric to describe feature importance "split " or is it "gain". In lightgbm we get to...

Aritro-94

indentify_zero_importance with asymmetrical data

10

Hi, Firstly I want to say thank you for such an amazing piece of work, really need this! Secondly, I am trying to identify_zero_importance features in my dataset, but when...

RustyBrain

AttributeError: 'FeatureSelector' object has no attribute 'data_all'

3

AttributeError: 'FeatureSelector' object has no attribute 'data_all'

yuanjie-ai

I have a dataset where the feature selector (identify_collinear) is identifying all of the features (3551 total features) are collinear. The output is: 3550 features with a correlation magnitude greater...

Rabia89

ERROR: Command "python setup.py egg_info" failed with error code 1

1

I tried to install it via pypi but I got below error; I use windows10 and python 3.7 (base) C:\Users\ilke.topak>pip install feature-selector Collecting feature-selector Using cached https://files.pythonhosted.org/packages/7c/d2/5448f8af6d3507f3c455429744c6436fdf3b91d10c75f4857d8e8bb4da1c/feature_selector-1.0.0-py3-none-any.whl Collecting lightgbm==2.1.1 (from...

ilkot

It doesn't seem that feature importances change. Using "lightgbm==2.3.0" I get the following; `xval, yval = make_classification(n_samples = 1000, n_features=10) model = lgb.LGBMClassifier(n_estimators=100, learning_rate = 0.05, verbose = -1) for...

psaks