Using embedder to transform categorical variables

embedder_transform.py

from embedder.regression import Embedder
from sklearn.pipeline import Pipeline
from xgboost import XGBRegressor

embedder = Embedder(embedding_dict)
embedder.fit(X_encoded, y)

X_embedded = embedder.transform(X)
# X_embedded = embedder.fit_transform(X, y)

# using sklearn pipeline
pipeline = Pipeline(
  [('embedding', Embedder(embedding_dict)),
   ('xgboost', XGBRegressor())]
)

pipeline.fit(X_encoded, y)
predictions = pipeline.predict(X_test, y_test)