微信扫一扫打赏支持

TensorFlow2_200729系列---12、全连接层

TensorFlow2_200729系列---12、全连接层

一、总结

一句话总结:

tf.keras.layers.Dense(2, activation='relu')

 

1、为什么神经节点一般有多个w和一个b?

A、h=relu(X@W)+b,这里的b表示bias,也就是误差,
B、比如第一层3个节点,第二层两个节点,对应的式子为:h1=x1*w1+x2*w2+x3*w3+b

 

 

2、tf.keras.layers.Dense(2, activation='relu') 中的dense表示什么意思?

dense表示连接的密度,也就是全连接,也就是所有的点都连上

 

 

3、tensorflow的model打印w和b?

for p in model.trainable_variables:print(p.name, p.shape)
# trainable_variables:
# Sequence of trainable variables owned by this module and its submodules.
for p in model.trainable_variables:
    print(p.name, p.shape)

dense_12/kernel:0 (3, 2)
dense_12/bias:0 (2,)
dense_13/kernel:0 (2, 2)
dense_13/bias:0 (2,)
dense_14/kernel:0 (2, 2)
dense_14/bias:0 (2,)

 

 

4、python 如何查看object有哪些属性和方法?

a、使用dir(object)或者object.__dict__
b、也可以使用help函数

 

 

 

二、全连接层

博客对应课程的视频位置:

 

 

 

 

 

import tensorflow as tf 
from     tensorflow import keras


x = tf.random.normal([2, 3])

model = keras.Sequential([
        keras.layers.Dense(2, activation='relu'),
        keras.layers.Dense(2, activation='relu'),
        keras.layers.Dense(2)
    ])
model.build(input_shape=[None, 3])
model.summary()

# trainable_variables:
# Sequence of trainable variables owned by this module and its submodules.
for p in model.trainable_variables:
    print(p.name, p.shape)

 

Model: "sequential_4"
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
dense_12 (Dense)             multiple                  8         
_________________________________________________________________
dense_13 (Dense)             multiple                  6         
_________________________________________________________________
dense_14 (Dense)             multiple                  6         
=================================================================
Total params: 20
Trainable params: 20
Non-trainable params: 0
_________________________________________________________________
dense_12/kernel:0 (3, 2)
dense_12/bias:0 (2,)
dense_13/kernel:0 (2, 2)
dense_13/bias:0 (2,)
dense_14/kernel:0 (2, 2)
dense_14/bias:0 (2,)

 

 

 

 

 
posted @ 2020-08-03 07:22  范仁义  阅读(154)  评论(0编辑  收藏  举报