def predat(numOtr, cls):
"""
random subset the dataset, then convert
to numpy format for further implementations
-------------------------------------------
input:
numOtr --total number of training samples
cls --number of classes
output:
sam_tr, ind_tr --numpy format training samples
sam_te, ind_te --numpy format testing samples
cls -- number of classes
"""
subset.main('dataset/YaleB.scale', numOtr, 0, 'dataset/YaleB.scale.tr', 'dataset/YaleB.scale.te')
lb_tr, ins_tr = svmutil.svm_read_problem('dataset/YaleB.scale.tr')
lb_te, ins_te = svmutil.svm_read_problem('dataset/YaleB.scale.te')
# change training data to numpy format
df = pd.DataFrame(ins_tr).fillna(0)
sam_tr = pd.DataFrame.as_matrix(df).T
ind_tr = np.array(lb_tr)
# change testing data to numpy format
df = pd.DataFrame(ins_te).fillna(0)
sam_te = pd.DataFrame.as_matrix(df).T
ind_te = np.array(lb_te)
return sam_tr, ind_tr, sam_te, ind_te, cls
def predictByFile(test_file_name, model):
y, X = svmutil.svm_read_problem(test_file_name)
p_label, p_acc, p_val = svmutil.svm_predict(y, X, model)
print len(p_label)
def trainByFile(train_file_name):
y, X = svmutil.svm_read_problem(train_file_name)
model = train(y, X)
return model
# -*- coding: utf-8 -*-
"""Excercise 6.2"""
import numpy as np
from libsvm.python import svm, svmutil
import os
dir = os.path.dirname(__file__)
abspath = os.path.join(dir, 'data.txt')
prob = svmutil.svm_read_problem(abspath)
print('Linear\n')
svmutil.svm_train(prob[0], prob[1], '-t 0')
print('RBF\n')
svmutil.svm_train(prob[0], prob[1], '-t 2')
print('converting to libsvm-format...')
gen_libsvm_format_data('dataset/X_train_precomputed.csv',
'dataset/Y_train.csv',
'train_precomputed.csv',
isKernel=True)
gen_libsvm_format_data('dataset/X_test_precomputed.csv',
'dataset/Y_test.csv',
'test_precomputed.csv',
isKernel=True)
gen_libsvm_format_data('dataset/X_train.csv', 'dataset/Y_train.csv',
'train.csv')
gen_libsvm_format_data('dataset/X_test.csv', 'dataset/Y_test.csv', 'test.csv')
#open training data
print('opening training data...')
y_train, x_train = svmutil.svm_read_problem('train.csv')
prob = svmutil.svm_problem(y_train, x_train)
y1, x_train_precomputed = svmutil.svm_read_problem('train_precomputed.csv')
prob_precomputed = svmutil.svm_problem(y1, x_train_precomputed, isKernel=True)
# grid search for the best parameters
print('grid searching...')
grid_search = {
'linear': '-t 0 -log2c -5,15,2',
'polynomial': '-t 1 -log2c -5,15,2 -log2g -5,15,2 -log2r -3,5,2 -d 4',
'RBF': '-t 2 -log2c -5,15,2 -log2g -5,15,2',
}
grid_search_precomputed = {'linear+RBF': '-t 4 -log2c -5,15,2'}
best_param = []
kernel_tab = []
def predictByFile( test_file_name, model ):
y, X = svmutil.svm_read_problem( test_file_name )
p_label, p_acc, p_val = svmutil.svm_predict(y, X, model)
print len(p_label)
def trainByFile( train_file_name ):
y, X = svmutil.svm_read_problem(train_file_name)
model = train(y, X)
return model
