def train_regression(training, params={}): # get data data,ds = transform_train_set(training) # generate regression m = rg.iter2matrix(data) x = m.T[:-1].T x = rg.append_ones(x) y = m.T[-1:].T w = rg.regression_weights(x,y) # get training error terror = rg.mse_error(x,y,w) return (w, ds, terror)
import unicodecsv as csv import numpy as np import scipy as sp import math import regression as rg dataf = open('data/features/feature_data.csv', 'rt') datac = csv.reader(dataf) next(datac) m = rg.iter2matrix(datac) x1 = m.T[:-1].T x = rg.append_ones(x1) y = m.T[-1:].T w = rg.regression_weights(x,y) lse = rg.lse_error(x, y, w) mse = rg.mse_error(x, y, w) print "weights: " + str(w) print "lse: " + str(lse) print "mse: " + str(mse)