用于 Scikit Learn 的 Keras 包装器 - AUC 评分器不起作用 [英] Keras Wrappers for Scikit Learn - AUC scorer is not working
问题描述
我正在尝试使用 Keras Scikit Learn Wrapper 来随机搜索参数更简单.我在这里写了一个示例代码:
I'm trying to use Keras Scikit Learn Wrapper in order to make random search for parameters easier. I wrote an example code here where :
- 我生成了一个人工数据集:
我正在使用 scikit learn
from sklearn.datasets import make_moons
dataset = make_moons(1000)
- 模型构建器定义:
我定义了需要的build_fn
函数:
def build_fn(nr_of_layers = 2,
first_layer_size = 10,
layers_slope_coeff = 0.8,
dropout = 0.5,
activation = "relu",
weight_l2 = 0.01,
act_l2 = 0.01,
input_dim = 2):
result_model = Sequential()
result_model.add(Dense(first_layer_size,
input_dim = input_dim,
activation=activation,
W_regularizer= l2(weight_l2),
activity_regularizer=activity_l2(act_l2)
))
current_layer_size = int(first_layer_size * layers_slope_coeff) + 1
for index_of_layer in range(nr_of_layers - 1):
result_model.add(BatchNormalization())
result_model.add(Dropout(dropout))
result_model.add(Dense(current_layer_size,
W_regularizer= l2(weight_l2),
activation=activation,
activity_regularizer=activity_l2(act_l2)
))
current_layer_size = int(current_layer_size * layers_slope_coeff) + 1
result_model.add(Dense(1,
activation = "sigmoid",
W_regularizer = l2(weight_l2)))
result_model.compile(optimizer="rmsprop", metrics = ["accuracy"], loss = "binary_crossentropy")
return result_model
NeuralNet = KerasClassifier(build_fn)
- 参数网格定义:
然后我定义了一个参数网格:
Then I defined a parameter grid :
param_grid = {
"nr_of_layers" : [2, 3, 4, 5],
"first_layer_size" : [5, 10, 15],
"layers_slope_coeff" : [0.4, 0.6, 0.8],
"dropout" : [0.3, 0.5, 0.8],
"weight_l2" : [0.01, 0.001, 0.0001],
"verbose" : [0],
"batch_size" : [1],
"nb_epoch" : [30]
}
- RandomizedSearchCV 阶段:
我定义了 RandomizedSearchCV
对象并拟合了来自人工数据集的值:
I defined RandomizedSearchCV
object and fitted with values from artificial dataset:
random_search = RandomizedSearchCV(NeuralNet,
param_distributions=param_grid, verbose=2, n_iter=1, scoring="roc_auc")
random_search.fit(dataset[0], dataset[1])
我得到的(在控制台中运行此代码后)是:
What I got (after running this code in console) is :
Traceback (most recent call last):
File "C:Anaconda2libsite-packagesIPythoncoreinteractiveshell.py", line 2885, in run_code
exec(code_obj, self.user_global_ns, self.user_ns)
File "<ipython-input-3-c5bdbc2770b7>", line 2, in <module>
random_search.fit(dataset[0], dataset[1])
File "C:Anaconda2libsite-packagessklearngrid_search.py", line 996, in fit
return self._fit(X, y, sampled_params)
File "C:Anaconda2libsite-packagessklearngrid_search.py", line 553, in _fit
for parameters in parameter_iterable
File "C:Anaconda2libsite-packagessklearnexternalsjoblibparallel.py", line 800, in __call__
while self.dispatch_one_batch(iterator):
File "C:Anaconda2libsite-packagessklearnexternalsjoblibparallel.py", line 658, in dispatch_one_batch
self._dispatch(tasks)
File "C:Anaconda2libsite-packagessklearnexternalsjoblibparallel.py", line 566, in _dispatch
job = ImmediateComputeBatch(batch)
File "C:Anaconda2libsite-packagessklearnexternalsjoblibparallel.py", line 180, in __init__
self.results = batch()
File "C:Anaconda2libsite-packagessklearnexternalsjoblibparallel.py", line 72, in __call__
return [func(*args, **kwargs) for func, args, kwargs in self.items]
File "C:Anaconda2libsite-packagessklearncross_validation.py", line 1550, in _fit_and_score
test_score = _score(estimator, X_test, y_test, scorer)
File "C:Anaconda2libsite-packagessklearncross_validation.py", line 1606, in _score
score = scorer(estimator, X_test, y_test)
File "C:Anaconda2libsite-packagessklearnmetricsscorer.py", line 175, in __call__
y_pred = y_pred[:, 1]
IndexError: index 1 is out of bounds for axis 1 with size 1
当我使用 accuracy
度量而不是使用 scoring = "roc_auc"
时,此代码工作正常.谁能解释一下我出了什么问题?有没有人遇到过类似的问题?
This code work fine when instead of using scoring = "roc_auc"
I used accuracy
metric. Can anyone explain me what's wrong? Have anyone had similiar problem?
推荐答案
KerasClassifier 中存在导致此问题的错误.我在 repo 上为它打开了一个问题.https://github.com/fchollet/keras/issues/2864
There is a bug in the KerasClassifier that is causing this issue. I have opened an issue for it on the repo. https://github.com/fchollet/keras/issues/2864
修复程序也在那里.您可以同时定义自己的 KerasClassifier 作为临时解决方法.
The fix is also in there. You can define your own KerasClassifier in the mean time as a temporary workaround.
class FixedKerasClassifier(KerasClassifier):
def predict_proba(self, X, **kwargs):
kwargs = self.filter_sk_params(Sequential.predict_proba, kwargs)
probs = self.model.predict_proba(X, **kwargs)
if(probs.shape[1] == 1):
probs = np.hstack([1-probs,probs])
return probs
这篇关于用于 Scikit Learn 的 Keras 包装器 - AUC 评分器不起作用的文章就介绍到这了,希望我们推荐的答案对大家有所帮助,也希望大家多多支持IT屋!