代码之家 › 专栏 › 技术社区 › Jonathan

keras中使用类权重的U-net自定义损失函数:三维目标不支持“类权重”

keras neural-network machine-learning

Jonathan · 技术社区 · 6 年前

inputs = Input((IMG_HEIGHT, IMG_WIDTH, IMG_CHANNELS))
...
outputs = Conv2D(4, (1, 1), activation='sigmoid') (c9)

model = Model(inputs=[inputs], outputs=[outputs])
model.compile(optimizer='adam', loss='dice', metrics=[mean_iou])

results = model.fit(X_train, Y_train, validation_split=0.1, batch_size=8, epochs=30, class_weight=class_weights)

我有四节课很不平衡。A类为70%,B类为15%,C类为10%,D类为5%。但是,我最关心的是D类,所以我做了以下计算: D_weight = A/D = 70/5 = 14 对于B类和A类的重量,以此类推。(如果有更好的方法来选择这些重量,请放心)

class_weights = {0: 1.0, 1: 6, 2: 7, 3: 14} .

但是,当我这样做时,我会得到以下错误。

class_weight

我有没有可能在最后一层之后添加一个密集层,并将其用作一个虚拟层,这样我就可以通过类权重,然后只使用最后一个conv2d层的输出来进行预测?

如果这不可能,我将如何修改loss函数(我知道 post

def dice_coef(y_true, y_pred):
    smooth = 1.
    y_true_f = K.flatten(y_true)
    y_pred_f = K.flatten(y_pred)
    intersection = K.sum(y_true_f * y_pred_f)
    return (2. * intersection + smooth) / (K.sum(y_true_f) + K.sum(y_pred_f) + smooth)

def bce_dice_loss(y_true, y_pred):
    return 0.5 * binary_crossentropy(y_true, y_pred) - dice_coef(y_true, y_pred)

但我看不出我能用什么方法输入类权重。如果有人想要完整的工作代码,请看这个 post . 但请记住将最后一个conv2d层的num类更改为4,而不是1。

1 回复 | 直到 5 年前

Daniel Möller 6 年前

你可以自己施加重量:

def weightedLoss(originalLossFunc, weightsList):

    def lossFunc(true, pred):

        axis = -1 #if channels last 
        #axis=  1 #if channels first


        #argmax returns the index of the element with the greatest value
        #done in the class axis, it returns the class index    
        classSelectors = K.argmax(true, axis=axis) 

        #considering weights are ordered by class, for each class
        #true(1) if the class index is equal to the weight index   
        classSelectors = [K.equal(i, classSelectors) for i in range(len(weightsList))]

        #casting boolean to float for calculations  
        #each tensor in the list contains 1 where ground true class is equal to its index 
        #if you sum all these, you will get a tensor full of ones. 
        classSelectors = [K.cast(x, K.floatx()) for x in classSelectors]

        #for each of the selections above, multiply their respective weight
        weights = [sel * w for sel,w in zip(classSelectors, weightsList)] 

        #sums all the selections
        #result is a tensor with the respective weight for each element in predictions
        weightMultiplier = weights[0]
        for i in range(1, len(weights)):
            weightMultiplier = weightMultiplier + weights[i]


        #make sure your originalLossFunc only collapses the class axis
        #you need the other axes intact to multiply the weights tensor
        loss = originalLossFunc(true,pred) 
        loss = loss * weightMultiplier

        return loss
    return lossFunc

直接在输入数据上更改类平衡

您也可以更改输入样本的平衡。

使用 `sample_weight`

你也可以“按样本”工作,而不是“按类”工作。

为输入数组中的每个样本创建一个权重数组: len(x_train) == len(weights)

以及 fit 将此数组传递给争论。
fit_generator ,发电机必须返回重量和列车/真对: return/yield inputs, targets, weights

keras中使用类权重的U-net自定义损失函数:三维目标不支持“类权重”

直接在输入数据上更改类平衡

使用 sample_weight

使用 `sample_weight`