解決Keras TensorFlow 混編中 trainable=False設(shè)置無效問題
這是最近碰到一個問題,先描述下問題:
首先我有一個訓(xùn)練好的模型(例如vgg16),我要對這個模型進(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來進(jìn)行模型優(yōu)化,tf的優(yōu)化器,默認(rèn)情況下對所有tf.trainable_variables()進(jìn)行權(quán)值更新,問題就出在這,明明將vgg16的模型設(shè)置為trainable=False,但是tf的優(yōu)化器仍然對vgg16做權(quán)值更新
以上就是問題描述,經(jīng)過谷歌百度等等,終于找到了解決辦法,下面我們一點(diǎn)一點(diǎn)的來復(fù)原整個問題。
trainable=False 無效
首先,我們導(dǎo)入訓(xùn)練好的模型vgg16,對其設(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)并沒有變化,也就是說設(shè)置無效
# 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)入模型的時候建立一個variable_scope,將需要訓(xùn)練的變量放在另一個variable_scope,然后通過tf.get_collection獲取需要訓(xùn)練的變量,最后通過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對于TensorFlow而言并不起作用
解決的辦法就是通過variable_scope對變量進(jìn)行區(qū)分,在通過tf.get_collection來獲取需要訓(xùn)練的變量,最后通過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)置無效問題就是小編分享給大家的全部內(nèi)容了,希望能給大家一個參考,也希望大家多多支持腳本之家。
相關(guān)文章
python 創(chuàng)建一個空dataframe 然后添加行數(shù)據(jù)的實例
今天小編就為大家分享一篇python 創(chuàng)建一個空dataframe 然后添加行數(shù)據(jù)的實例,具有很好的參考價值,希望對大家有所幫助。一起跟隨小編過來看看吧2018-06-06matplotlib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),文中通過示例代碼介紹的非常詳細(xì),對大家的學(xué)習(xí)或者工作具有一定的參考學(xué)習(xí)價值,需要的朋友們下面隨著小編來一起學(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
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,文中通過示例代碼介紹的非常詳細(xì),對大家的學(xué)習(xí)或者工作具有一定的參考學(xué)習(xí)價值,需要的朋友們下面隨著小編來一起學(xué)習(xí)學(xué)習(xí)吧2019-04-04Python Shiny庫創(chuàng)建交互式Web應(yīng)用及高級功能案例
Shiny是一個基于Python的交互式Web應(yīng)用框架,專注于簡化Web應(yīng)用的開發(fā)流程,本文將深入探討Shiny庫的基本用法、高級功能以及實際應(yīng)用案例,以幫助開發(fā)者充分發(fā)揮Shiny在Web應(yīng)用開發(fā)中的優(yōu)勢2023-12-12python經(jīng)典百題之static定義靜態(tài)變量的三種方法
日常腳本編寫過程中時常會用到python的靜態(tài)方法、實例方法、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2024-09-09python實現(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),具有一定的參考價值,感興趣的小伙伴們可以參考一下2019-02-02