Read and pre-process a CSV file for usage in a decision tree

The example uses the pruneabletree.csv_importer.CsvImporter to read a CSV file containing a data set with missing values and categorical attributes. Observations with missing values will be discarded and categorical attributes will be encoded. This makes the data ready to be used by a decision tree classifier.

print(__doc__)


from sklearn.model_selection import cross_val_score
from pruneabletree import CsvImporter, PruneableDecisionTreeClassifier

importer = CsvImporter()
clf = PruneableDecisionTreeClassifier(prune='ebp', ebp_confidence=0.001, random_state=0)

filepath = "pruneabletree/tests/datasets/dataset_56_vote.csv"
X, y = importer.fit_transform_both(filepath)
# Or:
# X = importer.fit_transform(filepath)
# y = importer.y
print(cross_val_score(clf, X, y, cv=10))

Total running time of the script: ( 0 minutes 0.000 seconds)