在 Keras 中使用 tf.metrics？

Question

我对specificity_at_sensitivity. Looking through the Keras docs特别感兴趣：

from keras import metrics

model.compile(loss='mean_squared_error',
              optimizer='sgd',
              metrics=[metrics.mae, metrics.categorical_accuracy])

但看起来 metrics 列表必须具有 arity 2 的函数，接受 (y_true, y_pred) 并返回单个张量值。

编辑：目前我是这样做的：

from sklearn.metrics import confusion_matrix

predictions = model.predict(x_test)
y_test = np.argmax(y_test, axis=-1)
predictions = np.argmax(predictions, axis=-1)
c = confusion_matrix(y_test, predictions)
print('Confusion matrix:\n', c)
print('sensitivity', c[0, 0] / (c[0, 1] + c[0, 0]))
print('specificity', c[1, 1] / (c[1, 1] + c[1, 0]))

这种方法的缺点是我只在训练完成后得到我关心的输出。更愿意每 10 个时期左右获取一次指标。

Answer 1

我认为对于只有两个传入参数没有严格的限制，在metrics.py函数中只有三个传入参数，但是k选择了默认值5。

def sparse_top_k_categorical_accuracy(y_true, y_pred, k=5):
    return K.mean(K.in_top_k(y_pred, K.cast(K.max(y_true, axis=-1), 'int32'), k), axis=-1)

Answer 2

我在 github, and it seems that tf.metrics are still not supported by Keras models. However, in case you are very interested in using tf.metrics.specificity_at_sensitivity, I would suggest the following workaround (inspired by BogdanRuzh's 解决方案中发现了相关问题):

def specificity_at_sensitivity(sensitivity, **kwargs):
    def metric(labels, predictions):
        # any tensorflow metric
        value, update_op = tf.metrics.specificity_at_sensitivity(labels, predictions, sensitivity, **kwargs)

        # find all variables created for this metric
        metric_vars = [i for i in tf.local_variables() if 'specificity_at_sensitivity' in i.name.split('/')[2]]

        # Add metric variables to GLOBAL_VARIABLES collection.
        # They will be initialized for new session.
        for v in metric_vars:
            tf.add_to_collection(tf.GraphKeys.GLOBAL_VARIABLES, v)

        # force to update metric values
        with tf.control_dependencies([update_op]):
            value = tf.identity(value)
            return value
    return metric


model.compile(loss='mean_squared_error',
              optimizer='sgd',
              metrics=[metrics.mae,
                       metrics.categorical_accuracy,
                       specificity_at_sensitivity(0.5)])

更新：

您可以使用 model.evaluate 在训练后检索指标。

在 Keras 中使用 tf.metrics？

Use tf.metrics in Keras?

python

tensorflow

keras

wrapper

keras-2