欧美bbbwbbbw肥妇,免费乱码人妻系列日韩,一级黄片

解決Keras TensorFlow 混編中 trainable=False設(shè)置無(wú)效問(wèn)題

 更新時(shí)間：2020年06月28日 14:40:13   作者：芥末的無(wú)奈  
這篇文章主要介紹了解決Keras TensorFlow 混編中 trainable=False設(shè)置無(wú)效問(wèn)題，具有很好的參考價(jià)值，希望對(duì)大家有所幫助。一起跟隨小編過(guò)來(lái)看看吧

這是最近碰到一個(gè)問(wèn)題，先描述下問(wèn)題：

首先我有一個(gè)訓(xùn)練好的模型(例如vgg16)，我要對(duì)這個(gè)模型進(jìn)行一些改變，例如添加一層全連接層，用于種種原因，我只能用TensorFlow來(lái)進(jìn)行模型優(yōu)化,tf的優(yōu)化器，默認(rèn)情況下對(duì)所有tf.trainable_variables()進(jìn)行權(quán)值更新，問(wèn)題就出在這，明明將vgg16的模型設(shè)置為trainable=False，但是tf的優(yōu)化器仍然對(duì)vgg16做權(quán)值更新

以上就是問(wèn)題描述，經(jīng)過(guò)谷歌百度等等，終于找到了解決辦法，下面我們一點(diǎn)一點(diǎn)的來(lái)復(fù)原整個(gè)問(wèn)題。

trainable=False 無(wú)效

首先，我們導(dǎo)入訓(xùn)練好的模型vgg16，對(duì)其設(shè)置成trainable=False

from keras.applications import VGG16
import tensorflow as tf
from keras import layers
# 導(dǎo)入模型
base_mode = VGG16(include_top=False)
# 查看可訓(xùn)練的變量
tf.trainable_variables()
[<tf.Variable 'block1_conv1/kernel:0' shape=(3, 3, 3, 64) dtype=float32_ref>,
 <tf.Variable 'block1_conv1/bias:0' shape=(64,) dtype=float32_ref>,
 <tf.Variable 'block1_conv2/kernel:0' shape=(3, 3, 64, 64) dtype=float32_ref>,
 <tf.Variable 'block1_conv2/bias:0' shape=(64,) dtype=float32_ref>,
 <tf.Variable 'block2_conv1/kernel:0' shape=(3, 3, 64, 128) dtype=float32_ref>,
 <tf.Variable 'block2_conv1/bias:0' shape=(128,) dtype=float32_ref>,
 <tf.Variable 'block2_conv2/kernel:0' shape=(3, 3, 128, 128) dtype=float32_ref>,
 <tf.Variable 'block2_conv2/bias:0' shape=(128,) dtype=float32_ref>,
 <tf.Variable 'block3_conv1/kernel:0' shape=(3, 3, 128, 256) dtype=float32_ref>,
 <tf.Variable 'block3_conv1/bias:0' shape=(256,) dtype=float32_ref>,
 <tf.Variable 'block3_conv2/kernel:0' shape=(3, 3, 256, 256) dtype=float32_ref>,
 <tf.Variable 'block3_conv2/bias:0' shape=(256,) dtype=float32_ref>,
 <tf.Variable 'block3_conv3/kernel:0' shape=(3, 3, 256, 256) dtype=float32_ref>,
 <tf.Variable 'block3_conv3/bias:0' shape=(256,) dtype=float32_ref>,
 <tf.Variable 'block4_conv1/kernel:0' shape=(3, 3, 256, 512) dtype=float32_ref>,
 <tf.Variable 'block4_conv1/bias:0' shape=(512,) dtype=float32_ref>,
 <tf.Variable 'block4_conv2/kernel:0' shape=(3, 3, 512, 512) dtype=float32_ref>,
 <tf.Variable 'block4_conv2/bias:0' shape=(512,) dtype=float32_ref>,
 <tf.Variable 'block4_conv3/kernel:0' shape=(3, 3, 512, 512) dtype=float32_ref>,
 <tf.Variable 'block4_conv3/bias:0' shape=(512,) dtype=float32_ref>,
 <tf.Variable 'block5_conv1/kernel:0' shape=(3, 3, 512, 512) dtype=float32_ref>,
 <tf.Variable 'block5_conv1/bias:0' shape=(512,) dtype=float32_ref>,
 <tf.Variable 'block5_conv2/kernel:0' shape=(3, 3, 512, 512) dtype=float32_ref>,
 <tf.Variable 'block5_conv2/bias:0' shape=(512,) dtype=float32_ref>,
 <tf.Variable 'block5_conv3/kernel:0' shape=(3, 3, 512, 512) dtype=float32_ref>,
 <tf.Variable 'block5_conv3/bias:0' shape=(512,) dtype=float32_ref>,
 <tf.Variable 'block1_conv1_1/kernel:0' shape=(3, 3, 3, 64) dtype=float32_ref>,
 <tf.Variable 'block1_conv1_1/bias:0' shape=(64,) dtype=float32_ref>,
 <tf.Variable 'block1_conv2_1/kernel:0' shape=(3, 3, 64, 64) dtype=float32_ref>,
 <tf.Variable 'block1_conv2_1/bias:0' shape=(64,) dtype=float32_ref>,
 <tf.Variable 'block2_conv1_1/kernel:0' shape=(3, 3, 64, 128) dtype=float32_ref>,
 <tf.Variable 'block2_conv1_1/bias:0' shape=(128,) dtype=float32_ref>,
 <tf.Variable 'block2_conv2_1/kernel:0' shape=(3, 3, 128, 128) dtype=float32_ref>,
 <tf.Variable 'block2_conv2_1/bias:0' shape=(128,) dtype=float32_ref>,
 <tf.Variable 'block3_conv1_1/kernel:0' shape=(3, 3, 128, 256) dtype=float32_ref>,
 <tf.Variable 'block3_conv1_1/bias:0' shape=(256,) dtype=float32_ref>,
 <tf.Variable 'block3_conv2_1/kernel:0' shape=(3, 3, 256, 256) dtype=float32_ref>,
 <tf.Variable 'block3_conv2_1/bias:0' shape=(256,) dtype=float32_ref>,
 <tf.Variable 'block3_conv3_1/kernel:0' shape=(3, 3, 256, 256) dtype=float32_ref>,
 <tf.Variable 'block3_conv3_1/bias:0' shape=(256,) dtype=float32_ref>,
 <tf.Variable 'block4_conv1_1/kernel:0' shape=(3, 3, 256, 512) dtype=float32_ref>,
 <tf.Variable 'block4_conv1_1/bias:0' shape=(512,) dtype=float32_ref>,
 <tf.Variable 'block4_conv2_1/kernel:0' shape=(3, 3, 512, 512) dtype=float32_ref>,
 <tf.Variable 'block4_conv2_1/bias:0' shape=(512,) dtype=float32_ref>,
 <tf.Variable 'block4_conv3_1/kernel:0' shape=(3, 3, 512, 512) dtype=float32_ref>,
 <tf.Variable 'block4_conv3_1/bias:0' shape=(512,) dtype=float32_ref>,
 <tf.Variable 'block5_conv1_1/kernel:0' shape=(3, 3, 512, 512) dtype=float32_ref>,
 <tf.Variable 'block5_conv1_1/bias:0' shape=(512,) dtype=float32_ref>,
 <tf.Variable 'block5_conv2_1/kernel:0' shape=(3, 3, 512, 512) dtype=float32_ref>,
 <tf.Variable 'block5_conv2_1/bias:0' shape=(512,) dtype=float32_ref>,
 <tf.Variable 'block5_conv3_1/kernel:0' shape=(3, 3, 512, 512) dtype=float32_ref>,
 <tf.Variable 'block5_conv3_1/bias:0' shape=(512,) dtype=float32_ref>]
# 設(shè)置 trainable=False
# base_mode.trainable = False似乎也是可以的
for layer in base_mode.layers:
  layer.trainable = False

設(shè)置好trainable=False后，再次查看可訓(xùn)練的變量，發(fā)現(xiàn)并沒(méi)有變化，也就是說(shuō)設(shè)置無(wú)效

# 再次查看可訓(xùn)練的變量
tf.trainable_variables()

[<tf.Variable 'block1_conv1/kernel:0' shape=(3, 3, 3, 64) dtype=float32_ref>,
 <tf.Variable 'block1_conv1/bias:0' shape=(64,) dtype=float32_ref>,
 <tf.Variable 'block1_conv2/kernel:0' shape=(3, 3, 64, 64) dtype=float32_ref>,
 <tf.Variable 'block1_conv2/bias:0' shape=(64,) dtype=float32_ref>,
 <tf.Variable 'block2_conv1/kernel:0' shape=(3, 3, 64, 128) dtype=float32_ref>,
 <tf.Variable 'block2_conv1/bias:0' shape=(128,) dtype=float32_ref>,
 <tf.Variable 'block2_conv2/kernel:0' shape=(3, 3, 128, 128) dtype=float32_ref>,
 <tf.Variable 'block2_conv2/bias:0' shape=(128,) dtype=float32_ref>,
 <tf.Variable 'block3_conv1/kernel:0' shape=(3, 3, 128, 256) dtype=float32_ref>,
 <tf.Variable 'block3_conv1/bias:0' shape=(256,) dtype=float32_ref>,
 <tf.Variable 'block3_conv2/kernel:0' shape=(3, 3, 256, 256) dtype=float32_ref>,
 <tf.Variable 'block3_conv2/bias:0' shape=(256,) dtype=float32_ref>,
 <tf.Variable 'block3_conv3/kernel:0' shape=(3, 3, 256, 256) dtype=float32_ref>,
 <tf.Variable 'block3_conv3/bias:0' shape=(256,) dtype=float32_ref>,
 <tf.Variable 'block4_conv1/kernel:0' shape=(3, 3, 256, 512) dtype=float32_ref>,
 <tf.Variable 'block4_conv1/bias:0' shape=(512,) dtype=float32_ref>,
 <tf.Variable 'block4_conv2/kernel:0' shape=(3, 3, 512, 512) dtype=float32_ref>,
 <tf.Variable 'block4_conv2/bias:0' shape=(512,) dtype=float32_ref>,
 <tf.Variable 'block4_conv3/kernel:0' shape=(3, 3, 512, 512) dtype=float32_ref>,
 <tf.Variable 'block4_conv3/bias:0' shape=(512,) dtype=float32_ref>,
 <tf.Variable 'block5_conv1/kernel:0' shape=(3, 3, 512, 512) dtype=float32_ref>,
 <tf.Variable 'block5_conv1/bias:0' shape=(512,) dtype=float32_ref>,
 <tf.Variable 'block5_conv2/kernel:0' shape=(3, 3, 512, 512) dtype=float32_ref>,
 <tf.Variable 'block5_conv2/bias:0' shape=(512,) dtype=float32_ref>,
 <tf.Variable 'block5_conv3/kernel:0' shape=(3, 3, 512, 512) dtype=float32_ref>,
 <tf.Variable 'block5_conv3/bias:0' shape=(512,) dtype=float32_ref>,
 <tf.Variable 'block1_conv1_1/kernel:0' shape=(3, 3, 3, 64) dtype=float32_ref>,
 <tf.Variable 'block1_conv1_1/bias:0' shape=(64,) dtype=float32_ref>,
 <tf.Variable 'block1_conv2_1/kernel:0' shape=(3, 3, 64, 64) dtype=float32_ref>,
 <tf.Variable 'block1_conv2_1/bias:0' shape=(64,) dtype=float32_ref>,
 <tf.Variable 'block2_conv1_1/kernel:0' shape=(3, 3, 64, 128) dtype=float32_ref>,
 <tf.Variable 'block2_conv1_1/bias:0' shape=(128,) dtype=float32_ref>,
 <tf.Variable 'block2_conv2_1/kernel:0' shape=(3, 3, 128, 128) dtype=float32_ref>,
 <tf.Variable 'block2_conv2_1/bias:0' shape=(128,) dtype=float32_ref>,
 <tf.Variable 'block3_conv1_1/kernel:0' shape=(3, 3, 128, 256) dtype=float32_ref>,
 <tf.Variable 'block3_conv1_1/bias:0' shape=(256,) dtype=float32_ref>,
 <tf.Variable 'block3_conv2_1/kernel:0' shape=(3, 3, 256, 256) dtype=float32_ref>,
 <tf.Variable 'block3_conv2_1/bias:0' shape=(256,) dtype=float32_ref>,
 <tf.Variable 'block3_conv3_1/kernel:0' shape=(3, 3, 256, 256) dtype=float32_ref>,
 <tf.Variable 'block3_conv3_1/bias:0' shape=(256,) dtype=float32_ref>,
 <tf.Variable 'block4_conv1_1/kernel:0' shape=(3, 3, 256, 512) dtype=float32_ref>,
 <tf.Variable 'block4_conv1_1/bias:0' shape=(512,) dtype=float32_ref>,
 <tf.Variable 'block4_conv2_1/kernel:0' shape=(3, 3, 512, 512) dtype=float32_ref>,
 <tf.Variable 'block4_conv2_1/bias:0' shape=(512,) dtype=float32_ref>,
 <tf.Variable 'block4_conv3_1/kernel:0' shape=(3, 3, 512, 512) dtype=float32_ref>,
 <tf.Variable 'block4_conv3_1/bias:0' shape=(512,) dtype=float32_ref>,
 <tf.Variable 'block5_conv1_1/kernel:0' shape=(3, 3, 512, 512) dtype=float32_ref>,
 <tf.Variable 'block5_conv1_1/bias:0' shape=(512,) dtype=float32_ref>,
 <tf.Variable 'block5_conv2_1/kernel:0' shape=(3, 3, 512, 512) dtype=float32_ref>,
 <tf.Variable 'block5_conv2_1/bias:0' shape=(512,) dtype=float32_ref>,
 <tf.Variable 'block5_conv3_1/kernel:0' shape=(3, 3, 512, 512) dtype=float32_ref>,
 <tf.Variable 'block5_conv3_1/bias:0' shape=(512,) dtype=float32_ref>]

解決的辦法

解決的辦法就是在導(dǎo)入模型的時(shí)候建立一個(gè)variable_scope，將需要訓(xùn)練的變量放在另一個(gè)variable_scope,然后通過(guò)tf.get_collection獲取需要訓(xùn)練的變量，最后通過(guò)tf的優(yōu)化器中var_list指定需要訓(xùn)練的變量

from keras import models
with tf.variable_scope('base_model'):
  base_model = VGG16(include_top=False, input_shape=(224,224,3))
with tf.variable_scope('xxx'):
  model = models.Sequential()
  model.add(base_model)
  model.add(layers.Flatten())
  model.add(layers.Dense(10))


# 獲取需要訓(xùn)練的變量
trainable_var = tf.get_collection(tf.GraphKeys.TRAINABLE_VARIABLES, 'xxx')
trainable_var

[<tf.Variable 'xxx_2/dense_1/kernel:0' shape=(25088, 10) dtype=float32_ref>,
<tf.Variable 'xxx_2/dense_1/bias:0' shape=(10,) dtype=float32_ref>]

# 定義tf優(yōu)化器進(jìn)行訓(xùn)練，這里假設(shè)有一個(gè)loss
loss = model.output / 2; # 隨便定義的，方便演示
train_step = tf.train.AdamOptimizer().minimize(loss, var_list=trainable_var)

總結(jié)

在keras與TensorFlow混編中，keras中設(shè)置trainable=False對(duì)于TensorFlow而言并不起作用

解決的辦法就是通過(guò)variable_scope對(duì)變量進(jìn)行區(qū)分，在通過(guò)tf.get_collection來(lái)獲取需要訓(xùn)練的變量，最后通過(guò)tf優(yōu)化器中var_list指定訓(xùn)練

以上這篇解決Keras TensorFlow 混編中 trainable=False設(shè)置無(wú)效問(wèn)題就是小編分享給大家的全部?jī)?nèi)容了，希望能給大家一個(gè)參考，也希望大家多多支持腳本之家。

相關(guān)文章

  • Python在cmd上打印彩色文字實(shí)現(xiàn)過(guò)程詳解

    Python在cmd上打印彩色文字實(shí)現(xiàn)過(guò)程詳解

    這篇文章主要介紹了Python在cmd上打印彩色文字實(shí)現(xiàn)過(guò)程詳解,文中通過(guò)示例代碼介紹的非常詳細(xì)，對(duì)大家的學(xué)習(xí)或者工作具有一定的參考學(xué)習(xí)價(jià)值,需要的朋友可以參考下
    2019-08-08
  • Python Numpy庫(kù)安裝與基本操作示例

    Python Numpy庫(kù)安裝與基本操作示例

    這篇文章主要介紹了Python Numpy庫(kù)安裝與基本操作,簡(jiǎn)單介紹了Numpy庫(kù)的基本功能、并結(jié)合實(shí)例形式分析了基于Numpy庫(kù)的數(shù)組與矩陣相關(guān)操作技巧,需要的朋友可以參考下
    2019-01-01
  • python 創(chuàng)建一個(gè)空dataframe 然后添加行數(shù)據(jù)的實(shí)例

    python 創(chuàng)建一個(gè)空dataframe 然后添加行數(shù)據(jù)的實(shí)例

    今天小編就為大家分享一篇python 創(chuàng)建一個(gè)空dataframe 然后添加行數(shù)據(jù)的實(shí)例，具有很好的參考價(jià)值，希望對(duì)大家有所幫助。一起跟隨小編過(guò)來(lái)看看吧
    2018-06-06
  • matplotlib命令與格式之tick坐標(biāo)軸日期格式(設(shè)置日期主副刻度)

    matplotlib命令與格式之tick坐標(biāo)軸日期格式(設(shè)置日期主副刻度)

    這篇文章主要介紹了matplotlib命令與格式之tick坐標(biāo)軸日期格式(設(shè)置日期主副刻度)，文中通過(guò)示例代碼介紹的非常詳細(xì)，對(duì)大家的學(xué)習(xí)或者工作具有一定的參考學(xué)習(xí)價(jià)值，需要的朋友們下面隨著小編來(lái)一起學(xué)習(xí)學(xué)習(xí)吧
    2019-08-08
  • 淺談python的輸入輸出，注釋，基本數(shù)據(jù)類型

    淺談python的輸入輸出，注釋，基本數(shù)據(jù)類型

    這篇文章主要介紹了python的輸入輸出，注釋，基本數(shù)據(jù)類型，文中通過(guò)示例代碼介紹的非常詳細(xì)，對(duì)大家的學(xué)習(xí)或者工作具有一定的參考學(xué)習(xí)價(jià)值，需要的朋友們下面隨著小編來(lái)一起學(xué)習(xí)學(xué)習(xí)吧
    2019-04-04
  • Python Shiny庫(kù)創(chuàng)建交互式Web應(yīng)用及高級(jí)功能案例

    Python Shiny庫(kù)創(chuàng)建交互式Web應(yīng)用及高級(jí)功能案例

    Shiny是一個(gè)基于Python的交互式Web應(yīng)用框架,專注于簡(jiǎn)化Web應(yīng)用的開(kāi)發(fā)流程,本文將深入探討Shiny庫(kù)的基本用法、高級(jí)功能以及實(shí)際應(yīng)用案例,以幫助開(kāi)發(fā)者充分發(fā)揮Shiny在Web應(yīng)用開(kāi)發(fā)中的優(yōu)勢(shì)
    2023-12-12
  • pyinstaller打包找不到文件的問(wèn)題解決

    pyinstaller打包找不到文件的問(wèn)題解決

    這篇文章主要介紹了pyinstaller打包找不到文件的問(wèn)題解決，文中通過(guò)示例代碼介紹的非常詳細(xì)，對(duì)大家的學(xué)習(xí)或者工作具有一定的參考學(xué)習(xí)價(jià)值，需要的朋友們下面隨著小編來(lái)一起學(xué)習(xí)學(xué)習(xí)吧
    2020-04-04
  • python經(jīng)典百題之static定義靜態(tài)變量的三種方法

    python經(jīng)典百題之static定義靜態(tài)變量的三種方法

    日常腳本編寫過(guò)程中時(shí)常會(huì)用到python的靜態(tài)方法、實(shí)例方法、類方法,下面這篇文章主要給大家介紹了關(guān)于python經(jīng)典百題之static定義靜態(tài)變量的三種方法,需要的朋友可以參考下
    2024-09-09
  • python3使用SMTP發(fā)送簡(jiǎn)單文本郵件

    python3使用SMTP發(fā)送簡(jiǎn)單文本郵件

    這篇文章主要為大家詳細(xì)介紹了python3使用SMTP發(fā)送簡(jiǎn)單文本郵件，文中示例代碼介紹的非常詳細(xì)，具有一定的參考價(jià)值，感興趣的小伙伴們可以參考一下
    2018-06-06
  • python實(shí)現(xiàn)學(xué)員管理系統(tǒng)

    python實(shí)現(xiàn)學(xué)員管理系統(tǒng)

    這篇文章主要為大家詳細(xì)介紹了python實(shí)現(xiàn)學(xué)員管理系統(tǒng)，文中示例代碼介紹的非常詳細(xì)，具有一定的參考價(jià)值，感興趣的小伙伴們可以參考一下
    2019-02-02

最新評(píng)論