def data_processing_linear_regression_with_mapping(filename, power):
white = pd.read_csv(filename, low_memory=False, sep=';').values
[N, d] = white.shape
maped_X = mapping_data(white[:, :-1], power)
print("MX DiM:", maped_X.shape)
white = np.insert(maped_X, maped_X.shape[1], white[:, -1], axis=1)
print("White Dim:", white.shape)
np.random.seed(3)
# prepare data
ridx = np.random.permutation(N)
ntr = int(np.round(N * 0.8))
nval = int(np.round(N * 0.1))
ntest = N - ntr - nval
# spliting training, validation, and test
Xtrain = np.hstack([np.ones([ntr, 1]), white[ridx[0:ntr], 0:-1]])
ytrain = white[ridx[0:ntr], -1]
Xval = np.hstack([np.ones([nval, 1]), white[ridx[ntr:ntr + nval], 0:-1]])
yval = white[ridx[ntr:ntr + nval], -1]
Xtest = np.hstack([np.ones([ntest, 1]), white[ridx[ntr + nval:], 0:-1]])
ytest = white[ridx[ntr + nval:], -1]
return Xtrain, ytrain, Xval, yval, Xtest, ytest
def data_processing_linear_regression(filename, non_invertible, mapping,
mapping_power):
white = pd.read_csv(filename, low_memory=False, sep=';').values
[N, d] = white.shape
if (mapping == True):
maped_X = mapping_data(white[:, :-1], mapping_power)
white = np.insert(maped_X, maped_X.shape[1], white[:, -1], axis=1)
np.random.seed(3)
# prepare data
ridx = np.random.permutation(N)
ntr = int(np.round(N * 0.8))
nval = int(np.round(N * 0.1))
ntest = N - ntr - nval
# spliting training, validation, and test
Xtrain = np.hstack([np.ones([ntr, 1]), white[ridx[0:ntr], 0:-1]])
ytrain = white[ridx[0:ntr], -1]
Xval = np.hstack([np.ones([nval, 1]), white[ridx[ntr:ntr + nval], 0:-1]])
yval = white[ridx[ntr:ntr + nval], -1]
Xtest = np.hstack([np.ones([ntest, 1]), white[ridx[ntr + nval:], 0:-1]])
# a = 0
# for i in Xval:
# for j in Xtrain:
# if (i == j).all():
# a +=1
# break
# print(len(Xtrain),len(Xval))
# print(a)
ytest = white[ridx[ntr + nval:], -1]
if (non_invertible == True):
N, D = Xtrain.shape
np.random.seed(4)
random_row = np.random.randint(N)
random_col = np.random.randint(D)
Xtrain[:, random_col] = 0
Xtrain[random_row, :] = 0
return Xtrain, ytrain, Xval, yval, Xtest, ytest
return Xtrain, ytrain, Xval, yval, Xtest, ytest
from data_loader import data_processing_linear_regression
from linear_regression import linear_regression_invertible
import numpy as np
filename = 'winequality-white.csv'
# Xtrain, ytrain, Xval, yval, Xtest, ytest = data_processing_linear_regression(filename, True, False, 0)
# w = linear_regression_invertible(Xtrain, ytrain)
# print('w is ', w)
from linear_regression import mapping_data
print(mapping_data(np.array([[1,2,3],[2,2,1]]), 3))
from linear_regression import mapping_data import numpy as np X = [1, 2, 3], [4, 5, 6] X = np.array(X) print(X) mapped_x = mapping_data(X, 2) print(mapped_x)
# obtain the index of the last element
end_idx = len(sample)
# print(end_idx)
# add that to the end of the original row
sample = np.insert(sample, end_idx, sample_power_i)
# print(sample.tolist())
# modify X
mapped_X[index] = sample
return np.asarray(mapped_X)
X=[[1,2,1],[2,1,2],[3,1,1]]
# print(X)
power = 3
Xtrain, ytrain, Xval, yval, Xtest, ytest = data_processing_linear_regression(filename, False, True, power)
print(mapping_data(Xtrain,1).shape)
print(mapping_data(Xtrain,power).shape)
print(mapping_data(X,1))
print(mapping_data(X,power))
# print(Y)
# Z = []
# for idx, row in enumerate(Y):
# c = [item for pair in zip(row, X3[idx]) for item in pair]
# Z.append(c)
# print(Z)
from linear_regression import mapping_data
import json
import numpy as np
import pandas as pd
white = pd.read_csv('winequality-white', low_memory=False, sep=';').values
[N, d] = white.shape
if (mapping == True):
maped_X = mapping_data(white[:, :-1], mapping_power)
white = np.insert(maped_X, maped_X.shape[1], white[:, -1], axis=1)
np.random.seed(3)
# prepare data
ridx = np.random.permutation(N)
ntr = int(np.round(N * 0.8))
nval = int(np.round(N * 0.1))
ntest = N - ntr - nval
# spliting training, validation, and test
Xtrain = np.hstack([np.ones([ntr, 1]), white[ridx[0:ntr], 0:-1]])
ytrain = white[ridx[0:ntr], -1]
Xval = np.hstack([np.ones([nval, 1]), white[ridx[ntr:ntr + nval], 0:-1]])
yval = white[ridx[ntr:ntr + nval], -1]
Xtest = np.hstack([np.ones([ntest, 1]), white[ridx[ntr + nval:], 0:-1]])
ytest = white[ridx[ntr + nval:], -1]
