首页 > 解决方案 > 如何使用分类焦点损失进行 keras 的一种热编码?

问题描述

我正在研究癫痫发作预测。我有不平衡的数据集,我想通过使用焦点损失使其平衡。我有 2 类 one-hot 编码向量。我找到了以下焦点损失代码,但我不知道如何才能y_pred在焦点损失代码中使用之前model.fit_generator

y_pred是模型的输出。那么在拟合我的模型之前,我如何在焦点损失代码中使用它?

焦点损失代码:

def categorical_focal_loss(gamma=2.0, alpha=0.25):
    """
    Implementation of Focal Loss from the paper in multiclass classification
    Formula:
        loss = -alpha*((1-p)^gamma)*log(p)
    Parameters:
        alpha -- the same as wighting factor in balanced cross entropy
        gamma -- focusing parameter for modulating factor (1-p)
    Default value:
        gamma -- 2.0 as mentioned in the paper
        alpha -- 0.25 as mentioned in the paper
    """
    def focal_loss(y_true, y_pred):
        # Define epsilon so that the backpropagation will not result in NaN
        # for 0 divisor case
        epsilon = K.epsilon()
        # Add the epsilon to prediction value
        #y_pred = y_pred + epsilon
        # Clip the prediction value
        y_pred = K.clip(y_pred, epsilon, 1.0-epsilon)
        # Calculate cross entropy
        cross_entropy = -y_true*K.log(y_pred)
        # Calculate weight that consists of  modulating factor and weighting factor
        weight = alpha * y_true * K.pow((1-y_pred), gamma)
        # Calculate focal loss
        loss = weight * cross_entropy
        # Sum the losses in mini_batch
        loss = K.sum(loss, axis=1)
        return loss
    
    return focal_loss

我的代码:

history=model.fit_generator(generate_arrays_for_training(indexPat, train_data, start=0,end=100)
validation_data=generate_arrays_for_training(indexPat, test_data, start=0,end=100)
steps_per_epoch=int((len(train_data)/2)), 
                                validation_steps=int((len(test_data)/2)),
                                verbose=2,epochs=65, max_queue_size=2, shuffle=True)
preictPrediction=model.predict_generator(generate_arrays_for_predict(indexPat, filesPath_data), max_queue_size=4, steps=len(filesPath_data))
y_pred1=np.argmax(preictPrediction,axis=1)
y_pred=list(y_pred1)

标签: pythontensorflowkerasloss-function

解决方案


为了社区的利益,来自评论部分。

这不是特定于focal loss,所有 keras 损失函数都采用 y_truey_pred,您不必担心这些参数来自哪里,它们是keras自动提供的。


推荐阅读