In machine learning, while working with scikit learn library, we need to save the trained models in a file and restore them in order to reuse it to compare the model with other models, to test the model on a new data. The saving of data is called Serializaion, while restoring the data is called Deserialization.

Also, we deal with different types and sizes of data. Some datasets are easily trained i.e- they take less time to train but the datasets whose size is large (more than 1GB) can take very large time to train on a local machine even with GPU. When we need the same trained data in some different project or later sometime, to avoid the wastage of the training time, store trained model so that it can be used anytime in the future.

There are two ways we can save a model in scikit learn:

1. Pickle string: The pickle module implements a fundamental, but powerful algorithm for serializing and de-serializing a Python object structure.

   Pickle model provides the following functions –

   
   

   
   

   pickle.dump to serialize an object hierarchy, you simply use dump().
   pickle.load to deserialize a data stream, you call the loads() function.

   Example: Let’s apply K Nearest Neighbor on iris dataset and then save the model.

   
   

   
   
   

   import numpy as np    # Load dataset from sklearn.datasets import load_iris iris = load_iris()    X = iris.data y = iris.target    # Split dataset into train and test X_train, X_test, y_train, y_test = \     train_test_split(X, y, test_size = 0.3,                         random_state = 2018)    # import KNeighborsClassifier model from sklearn.neighbors import KNeighborsClassifier as KNN knn = KNN(n_neighbors = 3)    # train model knn.fit(X_train, y_train)

   

   Save model to string using pickle –

   
   

   
   
   

   import pickle    # Save the trained model as a pickle string. saved_model = pickle.dumps(knn)    # Load the pickled model knn_from_pickle = pickle.loads(saved_model)    # Use the loaded pickled model to make predictions knn_from_pickle.predict(X_test)

   Output:
   

2. Pickled model as a file using joblib: Joblib is the replacement of pickle as it is more efficent on objects that carry large numpy arrays. These functions also accept file-like object instead of filenames.
   

   joblib.dump to serialize an object hierarchy
   joblib.load to deserialize a data stream

   Save to pickled file using joblib –

   
   

   
   
   

   from sklearn.externals import joblib    # Save the model as a pickle in a file joblib.dump(knn, 'filename.pkl')    # Load the model from the file knn_from_joblib = joblib.load('filename.pkl')     # Use the loaded model to make predictions knn_from_joblib.predict(X_test)

   Output: