import numpy as np
from utils.Answer import evaluation_test1
label = np.array([0, 1, 1, 1, 1, 0, 0, 1, 1, 0])
pred = np.array([0, 1, 0, 0, 1, 1, 1, 0, 0, 1])
hypo = np.array([0.43, 0.77, 0.31, 0.03, 0.52, 0.66, 0.90, 0.26, 0.44, 0.74])
result = evaluation_test1(label, pred, at=5)
real_result = {}
real_result['Accuracy '] = 0.300000
real_result['Precision'] = 0.400000
real_result['Recall '] = 0.333333
real_result['F_measure'] = 0.363636
for key in result.keys():
print('your: ', key, '\t:\t %.6f' % result[key], '\tanswer: ',
'\t:\t %.6f' % real_result[key])
# OPTIMIZER
OPTIMIZER = 'SGD'
assert DATA_NAME in ['Contracept', 'Heart', 'Yeast']
assert OPTIMIZER == 'SGD'
# Load dataset, model and evaluation metric
train_data, test_data, logistic_regression = _initialize(DATA_NAME)
train_x, train_y = train_data
num_data, num_features = train_x.shape
num_class = len(np.unique(train_y))
optim = SGD()
model = LogisticRegression(num_features)
loss, epoch = model.train(train_x, train_y, batch_size, num_epochs,
learning_rate, optim)
test_x, test_y = test_data
hypo, pred = model.predict(test_x)
pred = np.squeeze(pred)
# ======== Edit here =========
result = evaluation_test1(test_y, pred, at=90)
# =============================
for key in result.keys():
print(key, '\t\t\t:\t\t %.6f' % result[key])