1+ import pandas as pd
2+ import pickle
3+ import numpy as np
4+ from sklearn .neural_network import MLPRegressor
5+ from sklearn .datasets import make_regression
6+ 7+ NUM_FEATURES = 10
8+ NUM_SAMPLES = 1000
9+ 10+ # Make data
11+ def make_data ():
12+ X ,y = make_regression (n_samples = NUM_SAMPLES ,n_features = NUM_FEATURES ,
13+ n_informative = NUM_FEATURES ,noise = 0.5 )
14+ data = pd .DataFrame (X ,columns = ['X' + str (i ) for i in range (1 ,NUM_FEATURES + 1 )],dtype = np .float16 )
15+ data ['y' ]= np .array (y ,dtype = np .float16 )
16+ return data
17+ 18+ # Test/Train
19+ def test_train (data ):
20+ X_train ,y_train = data .iloc [:int (NUM_SAMPLES / 2 )].drop (['y' ],axis = 1 ),data .iloc [:int (NUM_SAMPLES / 2 )]['y' ]
21+ X_test ,y_test = data .iloc [int (NUM_SAMPLES / 2 ):].drop (['y' ],axis = 1 ),data .iloc [int (NUM_SAMPLES / 2 ):]['y' ]
22+ return (X_train ,y_train ,X_test ,y_test )
23+ 24+ # Fitting
25+ def fitting (X_train ,y_train ):
26+ mlp = MLPRegressor (max_iter = 50 )
27+ mlp .fit (X_train ,y_train )
28+ del X_train
29+ del y_train
30+ return mlp
31+ 32+ # Saving model
33+ def save (mlp ):
34+ with open ('MultiLayerPerceptron.sav' ,mode = 'wb' ) as f :
35+ pickle .dump (mlp ,f )
36+ 37+ def model_run (model ,testfile ):
38+ """
39+ Loads and runs a sklearn linear model
40+ """
41+ mlp = pickle .load (open (model , 'rb' ))
42+ X_test = pd .read_csv (testfile )
43+ _ = mlp .predict (X_test )
44+ return None
45+ 46+ if __name__ == '__main__' :
47+ data = make_data ()
48+ X_train ,y_train ,X_test ,y_test = test_train (data )
49+ X_test .to_csv ("Test.csv" ,index = False )
50+ mlp = fitting (X_train ,y_train )
51+ save (mlp )
52+ model_run ('MultiLayerPerceptron.sav' ,'Test.csv' )
0 commit comments