理解 tf.keras.layers.Dense()

huangapple go评论46阅读模式
英文:

Understanding tf.keras.layers.Dense()

问题

根据您提供的内容,以下是翻译好的部分:

我试图理解为什么直接计算密集层操作和使用`keras`实现之间存在差异。

根据文档(https://www.tensorflow.org/api_docs/python/tf/keras/layers/Dense),`tf.keras.layers.Dense()`应该实现操作`output = activation(dot(input, kernel) + bias)`,但下面的`result`和`result1`不相同。

```python
tf.random.set_seed(1)

bias = tf.Variable(tf.random.uniform(shape=(5,1)), dtype=tf.float32)
kernel = tf.Variable(tf.random.uniform(shape=(5,10)), dtype=tf.float32)
x = tf.constant(tf.random.uniform(shape=(10,1), dtype=tf.float32))

result = tf.nn.relu(tf.linalg.matmul(a=weights, b=x) + biases)
tf.print(result)

test = tf.keras.layers.Dense(units = 5, 
                            activation = 'relu',
                            use_bias = True, 
                            kernel_initializer = tf.keras.initializers.Constant(value=kernel), 
                            bias_initializer = tf.keras.initializers.Constant(value=bias), 
                            dtype=tf.float32)

result1 = test(tf.transpose(x))

print()
tf.print(result1)

输出

[[2.87080455]
 [3.25458574]
 [3.28776264]
 [3.14319134]
 [2.04760242]]

[[2.38769 3.63470697 2.62423944 3.31286287 2.91121125]]

使用test.get_weights()我可以看到内核和偏差(b)已设置为正确的值。我正在使用TF版本2.12.0。


<details>
<summary>英文:</summary>

I am trying to understand why there is a difference between calculating a dense layer operation directly and using the `keras` implementation. 

Following the documentation (https://www.tensorflow.org/api_docs/python/tf/keras/layers/Dense) `tf.keras.layers.Dense()` should implement the operation `output = activation(dot(input, kernel) + bias)` but `result` and `result1` below are not the same.

```python 
tf.random.set_seed(1)

bias = tf.Variable(tf.random.uniform(shape=(5,1)), dtype=tf.float32)
kernel = tf.Variable(tf.random.uniform(shape=(5,10)), dtype=tf.float32)
x = tf.constant(tf.random.uniform(shape=(10,1), dtype=tf.float32))

result = tf.nn.relu(tf.linalg.matmul(a=weights, b=x) + biases)
tf.print(result)

test = tf.keras.layers.Dense(units = 5, 
                            activation = &#39;relu&#39;,
                            use_bias = True, 
                            kernel_initializer = tf.keras.initializers.Constant(value=kernel), 
                            bias_initializer = tf.keras.initializers.Constant(value=bias), 
                            dtype=tf.float32)

result1 = test(tf.transpose(x))

print()
tf.print(result1)

output


[[2.87080455]
 [3.25458574]
 [3.28776264]
 [3.14319134]
 [2.04760242]]

[[2.38769 3.63470697 2.62423944 3.31286287 2.91121125]]

Using test.get_weights() I can see that the kernel and bias (b) are getting set to the correct values. I am using TF version 2.12.0.

答案1

得分: 0

经过一些实验,我意识到密集层的kernel需要具有shape=(10,5),而不是原始问题中的(5,10)。这是因为units=5,所以需要传递大小为10的向量(因此input_shape=(10,)被注释掉作为提醒)。以下是已校正的代码:

tf.random.set_seed(1)

bias   = tf.Variable(tf.random.uniform(shape=(5,1)), dtype=tf.float32)
kernel = tf.Variable(tf.random.uniform(shape=(10,5)), dtype=tf.float32)
x = tf.constant(tf.random.uniform(shape=(10,1), dtype=tf.float32))

result = tf.nn.relu(tf.linalg.matmul(a=weights, b=x, transpose_a=True) + biases)
tf.print(result)

test = tf.keras.layers.Dense(units = 5, 
                            # input_shape=(10,),
                            activation = 'relu',
                            use_bias = True, 
                            kernel_initializer = tf.keras.initializers.Constant(value=kernel), 
                            bias_initializer = tf.keras.initializers.Constant(value=bias), 
                            dtype=tf.float32)

result1 = test(tf.transpose(x))

print()
tf.print(result1)
[[2.38769]
 [3.63470697]
 [2.62423944]
 [3.31286287]
 [2.91121125]]

[[2.38769 3.63470697 2.62423944 3.31286287 2.91121125]]

最终,我不完全确定底层发生了什么以及为什么keras没有引发错误。我将检查tf.keras.layers.Dense()的实现,但任何已经了解代码的人的想法或建议都会非常赞赏!

英文:

After some experimentation I realized that the kernel for the dense layer needs to be of shape=(10,5) as apposed to (5,10) as in the code from the original question above. This is implicit because units=5 so a vector of size 10 needs to be passed (hence why input_shape=(10,) is commented out as a reminder). Below is the corrected code:

tf.random.set_seed(1)

bias   = tf.Variable(tf.random.uniform(shape=(5,1)), dtype=tf.float32)
kernel = tf.Variable(tf.random.uniform(shape=(10,5)), dtype=tf.float32)
x = tf.constant(tf.random.uniform(shape=(10,1), dtype=tf.float32))

result = tf.nn.relu(tf.linalg.matmul(a=weights, b=x, transpose_a=True) + biases)
tf.print(result)

test = tf.keras.layers.Dense(units = 5, 
                            # input_shape=(10,),
                            activation = &#39;relu&#39;,
                            use_bias = True, 
                            kernel_initializer = tf.keras.initializers.Constant(value=kernel), 
                            bias_initializer = tf.keras.initializers.Constant(value=bias), 
                            dtype=tf.float32)

result1 = test(tf.transpose(x))

print()
tf.print(result1)

[[2.38769]
 [3.63470697]
 [2.62423944]
 [3.31286287]
 [2.91121125]]

[[2.38769 3.63470697 2.62423944 3.31286287 2.91121125]]

Ultimately, I am not entirely sure what was happening under the hood and why keras did not raise an error. I will check with the tf.keras.layers.Dense() implementation but any thoughts or suggestions by someone who knows the code already are highly appreciated!

huangapple
  • 本文由 发表于 2023年5月24日 22:00:25
  • 转载请务必保留本文链接:https://go.coder-hub.com/76324368.html
匿名

发表评论

匿名网友

:?: :razz: :sad: :evil: :!: :smile: :oops: :grin: :eek: :shock: :???: :cool: :lol: :mad: :twisted: :roll: :wink: :idea: :arrow: :neutral: :cry: :mrgreen:

确定