def for_pca(data):
print "enter dimension to reduced to"
k = raw_input()
result = plot_pca(data,int(k))
i = 0
class_file = open("arcene_train.labels","r")
a = []
for line in class_file:
a.append(NP.append(result[i],[int(line)]))
i+=1
print a
nb.naive_base(a)
def for_pca(data):
print "enter dimension to reduced to"
k = raw_input()
result = plot_pca(data, int(k))
i = 0
class_file = open("arcene_train.labels", "r")
a = []
for line in class_file:
a.append(NP.append(result[i], [int(line)]))
i += 1
print a
nb.naive_base(a)
def for_lda(data):
class_file = open("arcene_train.labels","r")
a = []
i = 0
for line in class_file:
a.append((list(data[i]) + [int(line),]))
i+=1
print a
result = lda_main(NP.array(a))
i = 0
class_file.close()
class_file = open("arcene_train.labels","r")
a = []
for line in class_file:
a.append(list(result[i]) + [int(line)])
print a
nb.naive_base(NP.array(a))
def for_lda(data):
class_file = open("arcene_train.labels", "r")
a = []
i = 0
for line in class_file:
a.append((list(data[i]) + [
int(line),
]))
i += 1
print a
result = lda_main(NP.array(a))
i = 0
class_file.close()
class_file = open("arcene_train.labels", "r")
a = []
for line in class_file:
a.append(list(result[i]) + [int(line)])
print a
nb.naive_base(NP.array(a))
def cross_validation(k, dataset):
knn_result = 0
nb_result = 0
id3_result = 0
dataset_len = len(dataset)
for i in range(k):
x, y = int((i / k) * dataset_len), int(((i + 1) / k) * dataset_len)
test = dataset[x:y]
train = (dataset[:x] + (dataset[y:]))
knn_result += knn(train, test, 5)
nb_result += naive_base(train, test)
tree = tree_generator(train, attributes_dict)
id3_result += get_ID3_result(test, tree, attributes_dict)
save_accuracy("accuracy.txt", id3_result / k, knn_result / k, nb_result / k)
