fabulousdj/sortinghatlib

SortingHatLib

Library for ML feature type inference: https://github.com/pvn25/ML-Data-Prep-Zoo/tree/master/MLFeatureTypeInference

Original repo: https://github.com/pvn25/SortingHatLib.git

1. Install the package using python-pip

git clone https://github.com/fabulousdj/SortingHatLib.git

pip install SortingHatLib/

2. Import the library using

import sortinghat.pylib as pl

3. Read in csv file using pandas

# rf: Random Forest, neural: Neural Model, knn: K-nn, logreg: Logistic Regression, svm: RBF_SVM
dataDownstream = pd.read_csv('adult.csv')

4. Perform base featurization of the raw CSV file:

dataFeaturized = pl.FeaturizeFile(dataDownstream)

5. bigram feature extraction for Random Forest:

dataFeaturized1 = pl.FeatureExtraction(dataFeaturized)

6. Finally, load the model for prediction

y_RF = pl.Load_RF(dataFeaturized1)