如何在theano上实现加权二元交叉熵？-Java 学习之路

How to implement Weighted Binary CrossEntropy on theano?

我的卷积神经网络只能预测0~1（sigmoid） .

I want to penalize my predictions in this way :

Cost-Table

基本上，我想在模型预测为0时惩罚更多，但事实是1 .

Question : 如何使用theano和lasagne创建此加权二进制CrossEntropy函数？

I tried this below

prediction = lasagne.layers.get_output(model)


import theano.tensor as T
def weighted_crossentropy(predictions, targets):

    # Copy the tensor
    tgt = targets.copy("tgt")

    # Make it a vector
    # tgt = tgt.flatten()
    # tgt = tgt.reshape(3000)
    # tgt = tgt.dimshuffle(1,0)

    newshape = (T.shape(tgt)[0])
    tgt = T.reshape(tgt, newshape)

   #Process it so [index] < 0.5 = 0 , and [index] >= 0.5 = 1


    # Make it an integer.
    tgt = T.cast(tgt, 'int32')


    weights_per_label = theano.shared(lasagne.utils.floatX([0.2, 0.4]))

    weights = weights_per_label[tgt]  # returns a targets-shaped weight matrix
    loss = lasagne.objectives.aggregate(T.nnet.binary_crossentropy(predictions, tgt), weights=weights)

    return loss

loss_or_grads = weighted_crossentropy(prediction, self.target_var)

But I get this error below :

TypeError: New shape in reshape must be a vector or a list/tuple of scalar. Got Subtensor.0 after conversion to a vector.

参考：https://github.com/fchollet/keras/issues/2115

参考：https://groups.google.com/forum/#!topic/theano-users/R_Q4uG9BXp8

2 回答

感谢lasagne group的开发人员，我通过构建自己的损失函数来解决这个问题 .

loss_or_grads = -(customized_rate * target_var * tensor.log(prediction) + (1.0 - target_var) * tensor.log(1.0 - prediction))

loss_or_grads = loss_or_grads.mean()

回复于 2024-05-06T17:08:51+08:00

2
要解决语法错误：

更改
```
newshape = (T.shape(tgt)[0])
tgt = T.reshape(tgt, newshape)
```
至
```
newshape = (T.shape(tgt)[0],)
tgt = T.reshape(tgt, newshape)
```
T.reshape 期望一个轴元组，你没有提供这个，因此错误 .

在惩罚假阴性（预测0，真相1）之前，请确保此预测误差不是基于训练数据的统计数据，如@uyaseen suggested .
回复于 2024-05-06T17:08:51+08:00

如何在theano上实现加权二元交叉熵？

2 回答

相关问题