# Code from Chapter 9 of Machine Learning: An Algorithmic Perspective (2nd Edition)
# by Stephen Marsland (http://stephenmonika.net)

# You are free to use, change, or redistribute the code in any way you wish for
# non-commercial purposes, but please maintain the name of the original author.
# This code comes with no warranty of any kind.

# Stephen Marsland, 2008, 2014

# The iris classification example

def preprocessIris(infile,outfile):

    stext1 = 'Iris-setosa'
    stext2 = 'Iris-versicolor'
    stext3 = 'Iris-virginica'
    rtext1 = '0'
    rtext2 = '1'
    rtext3 = '2'

    fid = open(infile,"r")
    oid = open(outfile,"w")

    for s in fid:
        if s.find(stext1)>-1:
            oid.write(s.replace(stext1, rtext1))
        elif s.find(stext2)>-1:
            oid.write(s.replace(stext2, rtext2))
        elif s.find(stext3)>-1:
            oid.write(s.replace(stext3, rtext3))
    fid.close()
    oid.close()

import numpy as np
# Preprocessor to remove the test (only needed once)
#preprocessIris('/Users/srmarsla/Book/Datasets/Iris/iris.data','iris_proc.data')

iris = np.loadtxt('iris_proc.data',delimiter=',')
iris[:,:4] = iris[:,:4]-iris[:,:4].mean(axis=0)
imax = np.concatenate((iris.max(axis=0)*np.ones((1,5)),iris.min(axis=0)*np.ones((1,5))),axis=0).max(axis=0)
iris[:,:4] = iris[:,:4]/imax[:4]
#print iris[0:5,:]

# Split into training, validation, and test sets
target = np.zeros((np.shape(iris)[0],3));
indices = np.where(iris[:,4]==0) 
target[indices,0] = 1
indices = np.where(iris[:,4]==1)
target[indices,1] = 1
indices = np.where(iris[:,4]==2)
target[indices,2] = 1

# Randomly order the data
order = range(np.shape(iris)[0])
np.random.shuffle(order)
iris = iris[order,:]
target = target[order,:]

train = iris[::2,0:4]
traint = target[::2]
valid = iris[1::4,0:4]
validt = target[1::4]
test = iris[3::4,0:4]
testt = target[3::4]

#print train.max(axis=0), train.min(axis=0)

# Train the network
import mlp_cg
reload(mlp_cg)
net = mlp_cg.mlp_cg(train,traint,5,outtype='softmax')
net.confmat(test,testt)
net.mlptrain(train,traint)
net.confmat(test,testt)