Tensorflow 2.0：自定义 keras 指标导致 tf.function 回溯警告

Question

当我使用以下自定义指标（keras 样式）时：

from sklearn.metrics import classification_report, f1_score
from tensorflow.keras.callbacks import Callback

class Metrics(Callback):
    def __init__(self, dev_data, classifier, dataloader):
        self.best_f1_score = 0.0
        self.dev_data = dev_data
        self.classifier = classifier
        self.predictor = Predictor(classifier, dataloader)
        self.dataloader = dataloader

    def on_epoch_end(self, epoch, logs=None):
        print("start to evaluate....")
        _, preds = self.predictor(self.dev_data)
        y_trues, y_preds = [self.dataloader.label_vector(v["label"]) for v in self.dev_data], preds
        f1 = f1_score(y_trues, y_preds, average="weighted")
        print(classification_report(y_trues, y_preds,
                                    target_names=self.dataloader.vocab.labels))
        if f1 > self.best_f1_score:
            self.best_f1_score = f1
            self.classifier.save_model()
            print("best metrics, save model...")

我收到以下警告：

W1106 10:49:14.171694 4745115072 def_function.py:474] 6 out of the last 11 calls to .distributed_function at 0x14a3f9d90> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings is likely due to passing python objects instead of tensors. Also, tf.function has experimental_relax_shapes=True option that relaxes argument shapes that can avoid unnecessary retracing. Please refer to https://www.tensorflow.org/beta/tutorials/eager/tf_function#python_or_tensor_args and https://www.tensorflow.org/api_docs/python/tf/function for more details.

Answer 1

然后使用@tf.function(experimental_relax_shapes=True) 可能会解决您的问题

Answer 2

导入tensorflow后添加这一行：

tf.compat.v1.disable_eager_execution()

Answer 3

当回溯 TF 函数时会出现此警告，因为它的参数在形状或数据类型（对于张量）甚至值（Python 或 np 对象或变量）中发生变化。

在一般情况下，解决方法是在传递给 Keras 或 TF 某处的自定义函数定义之前使用 @tf.function(experimental_relax_shapes=True)。这会尝试检测并避免不必要的回溯，但不能保证解决问题。

在你的例子中，我猜 Predictor class 是一个自定义的 class，所以在 [= 的定义之前放置 @tf.function(experimental_relax_shapes=True) 18=]().

Tensorflow 2.0：自定义 keras 指标导致 tf.function 回溯警告

Tensorflow 2.0: custom keras metric caused tf.function retracing warning

python

python-3.x

tensorflow

tensorflow2.0