我最近从tensorflow切换到skflow.在tensorflow中,我们将lambda * tf.nn.l2_loss(weights)添加到我们的损失中.现在我在skflow中有以下代码：
def deep_psi(X, y):
layers = skflow.ops.dnn(X, [5, 10, 20, 10, 5], keep_prob=0.5)
preds, loss = skflow.models.logistic_regression(layers, y)
return preds, loss
def exp_decay(global_step):
return tf.train.exponential_decay(learning_rate=0.01,
global_step=global_step,
decay_steps=1000,
decay_rate=0.005)
deep_cd = skflow.TensorFlowEstimator(model_fn=deep_psi,
n_classes=2,
steps=10000,
batch_size=10,
learning_rate=exp_decay,
verbose=True,)
如何在此处以及在何处添加正则化器？ Illia暗示了here,但我不知道.
解决方法:
您仍然可以向损失中添加其他组件,只需要从dnn / logistic_regression中检索权重并将它们添加到损失中：
def regularize_loss(loss, weights, lambda):
for weight in weights:
loss = loss + lambda * tf.nn.l2_loss(weight)
return loss
def deep_psi(X, y):
layers = skflow.ops.dnn(X, [5, 10, 20, 10, 5], keep_prob=0.5)
preds, loss = skflow.models.logistic_regression(layers, y)
weights = []
for layer in range(5): # n layers you passed to dnn
weights.append(tf.get_variable("dnn/layer%d/linear/Matrix" % layer))
# biases are also available at dnn/layer%d/linear/Bias
weights.append(tf.get_variable('logistic_regression/weights'))
return preds, regularize_loss(loss, weights, lambda)
“`
注意,变量的路径可以是found here.
此外,我们想向所有具有变量的层(例如dnn,conv2d或fully_connected)添加正则化器支持,因此可能在下周晚上的Tensorflow构建中应具有类似dnn(..,regularize = tf.contrib.layers.l2_regularizer( lambda)).发生这种情况时,我将更新此答案.
