In [1]:

from IPython.core.display import HTML
css_file = 'style.css'
HTML(open(css_file, 'r').read())

Out[1]:

In [2]:

from numpy import array

In [3]:

import pickle

Unpickle a machine learning model¶

Reading the pickle¶

In [4]:

# Opening the pickle file
# The rb stands for read binary
model_pkl = open("Random_forest_regressor_model.pkl", "rb")

In [5]:

# Reading the model
model = pickle.load(model_pkl)

In [6]:

# Calling the model
model

Out[6]:

RandomForestRegressor(bootstrap=True, criterion='mse', max_depth=None,
           max_features='auto', max_leaf_nodes=None,
           min_impurity_split=1e-07, min_samples_leaf=1,
           min_samples_split=2, min_weight_fraction_leaf=0.0,
           n_estimators=100, n_jobs=-1, oob_score=True, random_state=42,
           verbose=0, warm_start=False)

Testing the model¶

In [7]:

# Confirming the number of features
model.n_features_

Out[7]:

In [8]:

# The importance of each feature
model.feature_importances_

Out[8]:

array([ 0.12339053,  0.10903826,  0.19797339,  0.45332269,  0.01259731,
        0.01452948,  0.01445595,  0.01312439,  0.01143903,  0.01015556,
        0.01254152,  0.01582894,  0.01160294])

In [9]:

# Testing the probability of a positive outcome of a new example
new_patient = array([[3, 16, 9, 22, 1, 0, 0, 0, 1, 0, 0, 1, 0]])
model.predict(new_patient)

Out[9]:

array([ 0.08])

In [ ]: