本文整理汇总了Python中Data.partition方法的典型用法代码示例。如果您正苦于以下问题：Python Data.partition方法的具体用法？Python Data.partition怎么用？Python Data.partition使用的例子？那么恭喜您, 这里精选的方法代码示例或许可以为您提供帮助。您也可以进一步了解该方法所在类Data的用法示例。
在下文中一共展示了Data.partition方法的2个代码示例，这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞，您的评价将有助于系统推荐出更棒的Python代码示例。
示例1: hyper_search
# 需要导入模块: import Data [as 别名]
# 或者: from Data import partition [as 别名]
def hyper_search(X, y, X_dev=None, y_dev=None, ccc=False):
if X_dev == None:
dataset = Data.partition(X, y)
X_train = dataset["X_train"]
y_train = dataset["y_train"]
X_dev = dataset["X_dev"]
y_dev = dataset["y_dev"]
else:
X_train = X
y_train = y
# Find Optimal Hyperparameter Setting
lam_arr = [0.01, 0.05, 0.1, 1, 10]
a_arr = [0, 0.001, 0.01, 0.1, 0.5, 1]
eta_arr = [0.01, 0.1, 0.3, 0.5, 0.7, 0.9, 1, 1.1]
n_arr = [10, 20, 30, 50]
var_prior_arr = [0.1, 0.5, 0.7, 1, 1.1, 1.5, 2]
# lam_arr = [0.02,0.03,0.04]
# a_arr = [0,0.01]
# eta_arr = [0.0001]
# n_arr = [50,100,150]
# var_prior_arr = [0.1,3]
best_ep = 1e6
best_em = 1e6
if ccc:
best_ep = -1e6
best_em = -1e6
params = {}
for lam in lam_arr:
for n in n_arr:
###############################FOR EP################################################
for var_prior in var_prior_arr:
err = ep_run(X_train, y_train, X_dev, y_dev, n, lam=lam, var_prior=var_prior, ccc=ccc)
if (ccc and err > best_ep) or (not ccc and err < best_ep):
best_ep = err
params["lam"] = lam
params["n"] = n
params["var_prior"] = var_prior
###############################FOR EM################################################
for eta in eta_arr:
for a in a_arr:
err = em_run(X_train, y_train, X_dev, y_dev, n, lam=lam, eta=eta, a=a, ccc=ccc)
if (ccc and err > best_em) or (not ccc and err < best_em):
best_em = err
params["lam_em"] = lam
params["n_em"] = n
params["eta"] = eta
params["a"] = a
# print params
print "best EP error: " + str(best_ep)
print "best EM error: " + str(best_em)
print "best params"
print params示例2: range
# 需要导入模块: import Data [as 别名]
# 或者: from Data import partition [as 别名]
###################### We load the word music dataset ###############################
# csv = np.genfromtxt ('music.csv', delimiter=",",skip_header=1)
# X = csv[ :, range(csv.shape[ 1 ] - 2) ]
# y = csv[ :, csv.shape[ 1 ] - 1 ]
#config.profile=True
################### We load power dataset ######################################
csv = np.genfromtxt ('power.csv', delimiter=",",skip_header=1)
X = csv[ 1:100, range(csv.shape[ 1 ] - 1) ]
y = csv[ 1:100, csv.shape[ 1 ] - 1 ]
dataset = Data.partition(X,y)
X_train = np.append(dataset['X_train'],dataset['X_dev'],axis=0)
y_train = np.append(dataset['y_train'],dataset['y_dev'],axis=0)
X_test = dataset['X_test']
y_test = dataset['y_test']
result = {'ep_train':0,'ep_test':0,
'em_train':0,'em_test':0,'svr':0}
for s in range(9):
# Find Optimal Hyperparameter Setting
np.random.seed(s)
r = EP_run.ep_run(X_train,y_train,X_test,y_test,n,lam=lam,var_prior=var_prior)
result['ep_train'] += r['train']
result['ep_test'] += r['test']
r = EM_run.em_run(X_train,y_train,X_test,y_test,n,lam=lam_em,eta=eta,a=a)
result['em_train'] += r['train']