14. Deployment - pickle

0. 개요

pickle type으로 모델을 저장한다 -> deployment

 

 

## Deployment ##
### Store Model for Later with Pickle ###
# import module
import pickle

# save the pipeline model to disk
pickle.dump(model, open('./model.pkl', 'wb'))

# load the pipeline model from disk and deserialize
model_load = pickle.load(open('./model.pkl', 'rb'))

# use the loaded pipeline model to predict on new data
y_pred = model_load.predict(X_test)