如何在 TensorFlow 2.0 中获取张量值？

Question

我尝试在 Keras 中创建自定义图层。我从例子中得到的这段代码。如何得到这一层的计算结果？

  class Linear(keras.layers.Layer):
        def __init__(self, units=32):
            super(Linear, self).__init__()
            self.units = units

        def build(self, input_shape):
            self.w = self.add_weight(
                shape=(input_shape[-1], self.units),
                initializer="random_normal",
                trainable=True,
            )
            self.b = self.add_weight(
                shape=(self.units,), initializer="random_normal", trainable=True
            )

        def call(self, inputs):
            return tf.matmul(inputs, self.w) + self.b


    x = tf.ones((2, 2))

    # At instantiation, we don't know on what inputs this is going to get called
    linear_layer = Linear(32)

    # The layer's weights are created dynamically the first time the layer is called
    y = linear_layer(x)

    print(type(y))
    print(y)

输出为：

<class 'tensorflow.python.framework.ops.Tensor'>
Tensor("linear_17/add:0", shape=(2, 32), dtype=float32)

我需要获取张量 y 的值。如何获得它们？ tf.Session 不工作，因为我使用 Tensorflow 2.2.0。

Answer 1

linear是一层，你要放到模型结构里

class Linear(keras.layers.Layer):
    def __init__(self, units=32):
        super(Linear, self).__init__()
        self.units = units

    def build(self, input_shape):
        self.w = self.add_weight(
            shape=(input_shape[-1], self.units),
            initializer="random_normal",
            trainable=True,
        )
        self.b = self.add_weight(
            shape=(self.units,), initializer="random_normal", trainable=True
        )

    def call(self, inputs):
        return tf.matmul(inputs, self.w) + self.b


x = tf.ones((2, 2))

inp = Input(2)
linear_layer = Linear(32)(inp)

model = Model(inp, linear_layer)
y = model.predict(x)

如何在 TensorFlow 2.0 中获取张量值？

How to get tensor value in Tensorflow 2.0?

keras

tensorflow

tensorflow2.0