for i in xrange(X.shape[0]):
dist = list()
for irow in range(self._parameter['neighbor_X'].shape[0]):
dist.append(np.linalg.norm(X[i, :] - self._parameter['neighbor_X'][irow, :]))
indices = np.argsort(dist)[:K]
pred.append(np.mean(self._parameter['neighbor_y'][indices]))
self._logger.info('progress: %.2f %%' % (float(i) / X.shape[0] * 100))
else:
raise ValueError
return pred
if __name__ == '__main__':
from base.time_scheduler import TimeScheduler
scheduler = TimeScheduler()
# KNN for classification task
path = os.getcwd() + '/../dataset/electricity-normalized.arff'
loader = DataLoader(path)
dataset = loader.load(target_col_name='class')
trainset, testset = dataset.cross_split()
knn = KNNClassifier(search_mode='kd_tree')
knn.fit(trainset[0], trainset[1])
predict_kd_tree = scheduler.tic_tac('kd_tree', knn.predict, X=testset[0])
knn = KNNClassifier(search_mode='brutal')
knn.fit(trainset[0], trainset[1])
predict_brutal = scheduler.tic_tac('brutal', knn.predict, X=testset[0])
scheduler.print_task_schedule('brutal')
scheduler.print_task_schedule('kd_tree')
print accuracy_score(testset[1], predict_brutal), accuracy_score(testset[1], predict_kd_tree)
dist.append(
np.linalg.norm(X[i, :] -
self._parameter['neighbor_X'][irow, :]))
indices = np.argsort(dist)[:K]
pred.append(np.mean(self._parameter['neighbor_y'][indices]))
self._logger.info('progress: %.2f %%' %
(float(i) / X.shape[0] * 100))
else:
raise ValueError
return pred
if __name__ == '__main__':
from base.time_scheduler import TimeScheduler
scheduler = TimeScheduler()
# KNN for classification task
path = os.getcwd() + '/../dataset/electricity-normalized.arff'
loader = DataLoader(path)
dataset = loader.load(target_col_name='class')
trainset, testset = dataset.cross_split()
knn = KNNClassifier(search_mode='kd_tree')
knn.fit(trainset[0], trainset[1])
predict_kd_tree = scheduler.tic_tac('kd_tree', knn.predict, X=testset[0])
knn = KNNClassifier(search_mode='brutal')
knn.fit(trainset[0], trainset[1])
predict_brutal = scheduler.tic_tac('brutal', knn.predict, X=testset[0])
scheduler.print_task_schedule('brutal')
scheduler.print_task_schedule('kd_tree')
print accuracy_score(testset[1], predict_brutal), accuracy_score(
for irow in range(self._parameter["neighbor_X"].shape[0]):
dist.append(np.linalg.norm(X[i, :] - self._parameter["neighbor_X"][irow, :]))
indices = np.argsort(dist)[:K]
pred.append(np.mean(self._parameter["neighbor_y"][indices]))
logger.info("progress: %.2f %%" % (float(i) / X.shape[0] * 100))
else:
raise ValueError
return pred
logger = get_logger("KNN")
if __name__ == "__main__":
from base.time_scheduler import TimeScheduler
scheduler = TimeScheduler()
# KNN for classification task
path = os.getcwd() + "/../dataset/electricity-normalized.arff"
loader = DataLoader(path)
dataset = loader.load(target_col_name="class")
trainset, testset = dataset.cross_split()
knn = KNNClassifier(search_mode="kd_tree")
knn.fit(trainset[0], trainset[1])
predict_kd_tree = scheduler.tic_tac("kd_tree", knn.predict, X=testset[0])
knn = KNNClassifier(search_mode="brutal")
knn.fit(trainset[0], trainset[1])
predict_brutal = scheduler.tic_tac("brutal", knn.predict, X=testset[0])
scheduler.print_task_schedule("brutal")
scheduler.print_task_schedule("kd_tree")
print accuracy_score(testset[1], predict_brutal), accuracy_score(testset[1], predict_kd_tree)
break
return np.array(result)
def _fast_sample_by_proba(self, sample_weight):
nArray = 1e5
cdf_map = list()
start_ix = 0
end_ix = 0
for i in xrange(self._N):
if i == self._N:
cdf_map.extend([i] * (self._N - start_ix + 1))
break
end_ix += int(math.floor(sample_weight[i] * nArray))
cdf_map.extend([i] * (end_ix - start_ix + 1))
start_ix = end_ix
cdf_map = np.array(cdf_map)
indices = np.random.randint(0, nArray, self._K)
return cdf_map[indices]
if __name__ == '__main__':
from base.time_scheduler import TimeScheduler
scheduler = TimeScheduler()
fast_sampler = Sampler(1e5, 1e3, mode='fast')
normal_sampler = Sampler(1e5, 1e3, mode='normal')
scheduler.tic_tac('sample_fast', fast_sampler.sample)
scheduler.tic_tac('sample_normal', normal_sampler.sample)
scheduler.print_task_schedule('sample_fast')
scheduler.print_task_schedule('sample_normal')
break
return np.array(result)
def _fast_sample_by_proba(self, sample_weight):
nArray = 1e5
cdf_map = list()
start_ix = 0
end_ix = 0
for i in xrange(self._N):
if i == self._N:
cdf_map.extend([i] * (self._N - start_ix + 1))
break
end_ix += int(math.floor(sample_weight[i] * nArray))
cdf_map.extend([i] * (end_ix - start_ix + 1))
start_ix = end_ix
cdf_map = np.array(cdf_map)
indices = np.random.randint(0, nArray, self._K)
return cdf_map[indices]
if __name__ == '__main__':
from base.time_scheduler import TimeScheduler
scheduler = TimeScheduler()
fast_sampler = Sampler(1e5, 1e3, mode='fast')
normal_sampler = Sampler(1e5, 1e3, mode='normal')
scheduler.tic_tac('sample_fast', fast_sampler.sample)
scheduler.tic_tac('sample_normal', normal_sampler.sample)
scheduler.print_task_schedule('sample_fast')
scheduler.print_task_schedule('sample_normal')
