random_num = random.sample(xrange(7070), 100)
random_data = []
for row in raw_data[1:]:
new_raw = []
for i in row[1:]:
if i in random_num:
new_raw.append(row[1:][i])
random_data.append(map(lambda x: float(x), new_raw[1:]))
# scale data for SVM
scaled_data = (preprocessing.MinMaxScaler()).fit_transform(data)
# feature selection
X_train = []
y_train = []
for i in range(len(scaled_data)):
if i not in [5, 8, 24, 28, 42, 54]:
X_train.append(scaled_data[i])
y_train.append(target[i])
X_test = list(scaled_data[i] for i in [5, 8, 24, 28, 42, 54])
y_test = list(target[i] for i in [5, 8, 24, 28, 42, 54])
lda = LDA()
lda.fit(X_train, y_train)
result = lda.predict(X_test)
print result
print y_test
print lda.get_params()
random_num = random.sample(xrange(7070), 100)
random_data = []
for row in raw_data[1:]:
new_raw = []
for i in row[1:]:
if i in random_num:
new_raw.append(row[1:][i])
random_data.append(map(lambda x: float(x), new_raw[1:]))
# scale data for SVM
scaled_data = (preprocessing.MinMaxScaler()).fit_transform(data)
# feature selection
X_train = []
y_train = []
for i in range(len(scaled_data)):
if i not in [5, 8, 24, 28, 42, 54]:
X_train.append(scaled_data[i])
y_train.append(target[i])
X_test = list(scaled_data[i] for i in [5, 8, 24, 28, 42, 54])
y_test = list(target[i] for i in [5, 8, 24, 28, 42, 54])
lda = LDA()
lda.fit(X_train, y_train)
result = lda.predict(X_test)
print result
print y_test
print lda.get_params()
