slim.get_model_variables()

原创于 2022-09-15 14:43:53 发布 · 1.8k 阅读

0 ·

CC 4.0 BY-SA版权

文章标签：

#tensorflow #人工智能 #python

Slim 专栏收录该内容

16 篇文章

订阅专栏

该博客介绍了 TensorFlow 中的 `get_model_variables` 函数，这是一个用于获取模型变量的实用工具。它允许用户根据作用域和后缀来筛选变量。参数包括 `scope`，用于指定变量的作用域，以及 `suffix`，用于过滤具有特定后缀的变量。返回值是一个包含匹配条件的变量列表，帮助开发者更方便地管理和操作模型参数。

参考 slim.get_model_variables() - 云+社区 - 腾讯云

def get_model_variables(scope=None, suffix=None):

  return get_variables(scope, suffix, ops.GraphKeys.MODEL_VARIABLES)

获取按范围和/或后缀过滤的模型变量列表。

参数：

scope:筛选要返回的变量的可选作用域
suffix:用于过滤要返回的变量的可选后缀

返回值：

集合中具有范围和后缀的变量列表

确定要放弃本次机会？

福利倒计时

: :

立减 ¥

普通VIP年卡可用

立即使用

Wanderer001

关注关注

0
点赞
踩
0

收藏

觉得还不错? 一键收藏
0
评论
分享

复制链接

分享到 QQ

分享到新浪微博

扫一扫
打赏
打赏
打赏举报

举报

专栏目录

TF版FasterRCNN：resnet_v1.py代码阅读笔记

gusui7202的博客

12-04

4076

个人代码阅读笔记。第二次更新：2019.4.3 # -------------------------------------------------------- # Tensorflow Faster R-CNN # Licensed under The MIT License [see LICENSE for details] # Written by Zheqi He and Xi...

TensorFlow从入门到放弃（二）——基于InceptionV3的迁移学习以及图像特征的提取

澎湃de家夥的博客

07-20

6196

1. flower数据集下载地址：http://download.tensorflow.org/example_images/flower_photos.tgz 共五种花的图片 2. 图片处理将图片划分为train、val、test三个子集并提取图片特征。这个过程有点儿漫长请耐心等待。。。。。。 import glob import os.path import numpy a...

参与评论您还未登录，请先登录后发表或查看评论

Tensor flow.contrib.slim例子之slim.getvariables()

猫头军师兵书

11-30

1665

import tensorflow as tf import tensorflow.contrib.slim as slim # Model Variables weights = slim.model_variable('weights', shape=[5, 5, 3 , 3], ...

java源码修炼（三） getModel()

封的博客

11-08

5985

ClientaaaaQuery aaaaQuery = getModel(ClientaaaaQuery.class); 其实这个getModel（）方法就是要得到类的实例，这个类ClientaaaQuery中的字段对应页面上的字段。这种方法和SpringMVC得到实体类一样。贴一段代码： public T getModel(Class modelClass) {

pythonmessage用法_Python Message.Message方法代碼示例

weixin_39588265的博客

11-29

2232

本文整理匯總了Python中email.Message.Message方法的典型用法代碼示例。如果您正苦於以下問題：Python Message.Message方法的具體用法？Python Message.Message怎麽用？Python Message.Message使用的例子？那麽恭喜您, 這裏精選的方法代碼示例或許可以為您提供幫助。您也可以進一步了解該方法所在模塊email.Message...

tensorflow中的slim函数集合

weixin_36670529的博客

09-11

1428

目录 slim.get_model_variables() slim.get_trainable_variables() slim.learning.train() slim.fully_connected() slim.softmax() slim.get_model_variables() slim.get_or_create_global_step() slim.arg_sc...

Tensorflow-slim 学习笔记（一）

_提拉米苏的博客

01-26

882

Tensorflow-slim 学习笔记（一）接触到slim之后，了解到其较为轻量级的一个库，那么就是我们这篇文章要叙述的了，slim" role="presentation" style="position: relative;">slimslimslim。首先我们把github上的地址贴出来： https://github.com/tensorflow/tensorflow/tree/m

def generator(self, inputs, reuse=False, scope='g_net'): n, h, w, c = inputs.get_shape().as_list() if self.args.model == 'lstm': with tf.variable_scope('LSTM'): cell = BasicConvLSTMCell([h / 4, w / 4], [3, 3], 128) rnn_state = cell.zero_state(batch_size=self.batch_size, dtype=tf.float32) x_unwrap = [] with tf.variable_scope(scope, reuse=reuse): with slim.arg_scope([slim.conv2d, slim.conv2d_transpose], activation_fn=tf.nn.relu, padding='SAME', normalizer_fn=None, weights_initializer=tf.contrib.layers.xavier_initializer(uniform=True), biases_initializer=tf.constant_initializer(0.0)): inp_pred = inputs for i in xrange(self.n_levels): scale = self.scale ** (self.n_levels - i - 1) hi = int(round(h * scale)) wi = int(round(w * scale)) inp_blur = tf.image.resize_images(inputs, [hi, wi], method=0) inp_pred = tf.stop_gradient(tf.image.resize_images(inp_pred, [hi, wi], method=0)) inp_all = tf.concat([inp_blur, inp_pred], axis=3, name='inp') if self.args.model == 'lstm': rnn_state = tf.image.resize_images(rnn_state, [hi // 4, wi // 4], method=0) # encoder conv1_1 = slim.conv2d(inp_all, 32, [5, 5], scope='enc1_1') conv1_2 = ResnetBlock(conv1_1, 32, 5, scope='enc1_2') conv1_3 = ResnetBlock(conv1_2, 32, 5, scope='enc1_3') conv1_4 = ResnetBlock(conv1_3, 32, 5, scope='enc1_4') conv2_1 = slim.conv2d(conv1_4, 64, [5, 5], stride=2, scope='enc2_1') conv2_2 = ResnetBlock(conv2_1, 64, 5, scope='enc2_2') conv2_3 = ResnetBlock(conv2_2, 64, 5, scope='enc2_3') conv2_4 = ResnetBlock(conv2_3, 64, 5, scope='enc2_4') conv3_1 = slim.conv2d(conv2_4, 128, [5, 5], stride=2, scope='enc3_1') conv3_2 = ResnetBlock(conv3_1, 128, 5, scope='enc3_2') conv3_3 = ResnetBlock(conv3_2, 128, 5, scope='enc3_3') conv3_4 = ResnetBlock(conv3_3, 128, 5, scope='enc3_4') if self.args.model == 'lstm': deconv3_4, rnn_state = cell(conv3_4, rnn_state) else: deconv3_4 = conv3_4 # decoder deconv3_3 = ResnetBlock(deconv3_4, 128, 5, scope='dec3_3') deconv3_2 = ResnetBlock(deconv3_3, 128, 5, scope='dec3_2') deconv3_1 = ResnetBlock(deconv3_2, 128, 5, scope='dec3_1') deconv2_4 = slim.conv2d_transpose(deconv3_1, 64, [4, 4], stride=2, scope='dec2_4') cat2 = deconv2_4 + conv2_4 deconv2_3 = ResnetBlock(cat2, 64, 5, scope='dec2_3') deconv2_2 = ResnetBlock(deconv2_3, 64, 5, scope='dec2_2') deconv2_1 = ResnetBlock(deconv2_2, 64, 5, scope='dec2_1') deconv1_4 = slim.conv2d_transpose(deconv2_1, 32, [4, 4], stride=2, scope='dec1_4') cat1 = deconv1_4 + conv1_4 deconv1_3 = ResnetBlock(cat1, 32, 5, scope='dec1_3') deconv1_2 = ResnetBlock(deconv1_3, 32, 5, scope='dec1_2') deconv1_1 = ResnetBlock(deconv1_2, 32, 5, scope='dec1_1') inp_pred = slim.conv2d(deconv1_1, self.chns, [5, 5], activation_fn=None, scope='dec1_0') if i >= 0: x_unwrap.append(inp_pred) if i == 0: tf.get_variable_scope().reuse_variables() return x_unwrap

03-23

tf.get_variable_scope().reuse_variables() ``` - **输出列表**：`x_unwrap`保存每个尺度的预测结果。 - **变量重用**：从第二层级开始共享参数（减少计算量）。 --- ### 8. **应用场景** - **图像超分辨率**：...

x_unwrap = [] with tf.variable_scope(scope, reuse=reuse): with slim.arg_scope([slim.conv2d, slim.conv2d_transpose], activation_fn=tf.nn.relu, padding='SAME', normalizer_fn=None, weights_initializer=tf.contrib.layers.xavier_initializer(uniform=True), biases_initializer=tf.constant_initializer(0.0)): inp_pred = inputs for i in xrange(self.n_levels): scale = self.scale ** (self.n_levels - i - 1) hi = int(round(h * scale)) wi = int(round(w * scale)) inp_blur = tf.image.resize_images(inputs, [hi, wi], method=0) inp_pred = tf.stop_gradient(tf.image.resize_images(inp_pred, [hi, wi], method=0)) inp_all = tf.concat([inp_blur, inp_pred], axis=3, name='inp') if self.args.model == 'lstm': rnn_state = tf.image.resize_images(rnn_state, [hi // 4, wi // 4], method=0) # encoder conv1_1 = slim.conv2d(inp_all, 32, [5, 5], scope='enc1_1') conv1_2 = ResnetBlock(conv1_1, 32, 5, scope='enc1_2') conv1_3 = ResnetBlock(conv1_2, 32, 5, scope='enc1_3') conv1_4 = ResnetBlock(conv1_3, 32, 5, scope='enc1_4') conv2_1 = slim.conv2d(conv1_4, 64, [5, 5], stride=2, scope='enc2_1') conv2_2 = ResnetBlock(conv2_1, 64, 5, scope='enc2_2') conv2_3 = ResnetBlock(conv2_2, 64, 5, scope='enc2_3') conv2_4 = ResnetBlock(conv2_3, 64, 5, scope='enc2_4') conv3_1 = slim.conv2d(conv2_4, 128, [5, 5], stride=2, scope='enc3_1') conv3_2 = ResnetBlock(conv3_1, 128, 5, scope='enc3_2') conv3_3 = ResnetBlock(conv3_2, 128, 5, scope='enc3_3') conv3_4 = ResnetBlock(conv3_3, 128, 5, scope='enc3_4') if self.args.model == 'lstm': deconv3_4, rnn_state = cell(conv3_4, rnn_state) else: deconv3_4 = conv3_4 # decoder deconv3_3 = ResnetBlock(deconv3_4, 128, 5, scope='dec3_3') deconv3_2 = ResnetBlock(deconv3_3, 128, 5, scope='dec3_2') deconv3_1 = ResnetBlock(deconv3_2, 128, 5, scope='dec3_1') deconv2_4 = slim.conv2d_transpose(deconv3_1, 64, [4, 4], stride=2, scope='dec2_4') cat2 = deconv2_4 + conv2_4 deconv2_3 = ResnetBlock(cat2, 64, 5, scope='dec2_3') deconv2_2 = ResnetBlock(deconv2_3, 64, 5, scope='dec2_2') deconv2_1 = ResnetBlock(deconv2_2, 64, 5, scope='dec2_1') deconv1_4 = slim.conv2d_transpose(deconv2_1, 32, [4, 4], stride=2, scope='dec1_4') cat1 = deconv1_4 + conv1_4 deconv1_3 = ResnetBlock(cat1, 32, 5, scope='dec1_3') deconv1_2 = ResnetBlock(deconv1_3, 32, 5, scope='dec1_2') deconv1_1 = ResnetBlock(deconv1_2, 32, 5, scope='dec1_1') inp_pred = slim.conv2d(deconv1_1, self.chns, [5, 5], activation_fn=None, scope='dec1_0') if i >= 0: x_unwrap.append(inp_pred) if i == 0: tf.get_variable_scope().reuse_variables() return x_unwrap 下面这段代码是什么意思if i >= 0: x_unwrap.append(inp_pred) if i == 0: tf.get_variable_scope().reuse_variables()

最新发布

03-26

例如，代码中有一个`tf.get_variable_scope().reuse_variables()`在i=0时被调用，那么当i=1,2,...时，变量作用域已经处于复用状态，此时创建同名的变量会引发错误，除非它们已经被创建过。但根据代码结构，每次循环...

with tf.variable_scope(scope, reuse=reuse): with slim.arg_scope([slim.conv2d, slim.conv2d_transpose], activation_fn=tf.nn.relu, padding='SAME', normalizer_fn=None, weights_initializer=tf.contrib.layers.xavier_initializer(uniform=True), biases_initializer=tf.constant_initializer(0.0)): inp_pred = inputs for i in xrange(self.n_levels): scale = self.scale ** (self.n_levels - i - 1) hi = int(round(h * scale)) wi = int(round(w * scale)) inp_blur = tf.image.resize_images(inputs, [hi, wi], method=0) inp_pred = tf.stop_gradient(tf.image.resize_images(inp_pred, [hi, wi], method=0)) inp_all = tf.concat([inp_blur, inp_pred], axis=3, name='inp') if self.args.model == 'lstm': rnn_state = tf.image.resize_images(rnn_state, [hi // 4, wi // 4], method=0) # encoder conv1_1 = slim.conv2d(inp_all, 32, [5, 5], scope='enc1_1') conv1_2 = ResnetBlock(conv1_1, 32, 5, scope='enc1_2') conv1_3 = ResnetBlock(conv1_2, 32, 5, scope='enc1_3') conv1_4 = ResnetBlock(conv1_3, 32, 5, scope='enc1_4') conv2_1 = slim.conv2d(conv1_4, 64, [5, 5], stride=2, scope='enc2_1') conv2_2 = ResnetBlock(conv2_1, 64, 5, scope='enc2_2') conv2_3 = ResnetBlock(conv2_2, 64, 5, scope='enc2_3') conv2_4 = ResnetBlock(conv2_3, 64, 5, scope='enc2_4') conv3_1 = slim.conv2d(conv2_4, 128, [5, 5], stride=2, scope='enc3_1') conv3_2 = ResnetBlock(conv3_1, 128, 5, scope='enc3_2') conv3_3 = ResnetBlock(conv3_2, 128, 5, scope='enc3_3') conv3_4 = ResnetBlock(conv3_3, 128, 5, scope='enc3_4') if self.args.model == 'lstm': deconv3_4, rnn_state = cell(conv3_4, rnn_state) else: deconv3_4 = conv3_4 # decoder deconv3_3 = ResnetBlock(deconv3_4, 128, 5, scope='dec3_3') deconv3_2 = ResnetBlock(deconv3_3, 128, 5, scope='dec3_2') deconv3_1 = ResnetBlock(deconv3_2, 128, 5, scope='dec3_1') deconv2_4 = slim.conv2d_transpose(deconv3_1, 64, [4, 4], stride=2, scope='dec2_4') cat2 = deconv2_4 + conv2_4 deconv2_3 = ResnetBlock(cat2, 64, 5, scope='dec2_3') deconv2_2 = ResnetBlock(deconv2_3, 64, 5, scope='dec2_2') deconv2_1 = ResnetBlock(deconv2_2, 64, 5, scope='dec2_1') deconv1_4 = slim.conv2d_transpose(deconv2_1, 32, [4, 4], stride=2, scope='dec1_4') cat1 = deconv1_4 + conv1_4 deconv1_3 = ResnetBlock(cat1, 32, 5, scope='dec1_3') deconv1_2 = ResnetBlock(deconv1_3, 32, 5, scope='dec1_2') deconv1_1 = ResnetBlock(deconv1_2, 32, 5, scope='dec1_1') inp_pred = slim.conv2d(deconv1_1, self.chns, [5, 5], activation_fn=None, scope='dec1_0') if i >= 0: x_unwrap.append(inp_pred) if i == 0: tf.get_variable_scope().reuse_variables() return x_unwrap

03-23

tf.get_variable_scope().reuse_variables() # 首次循环后启用变量重用 ``` - $\text{最终卷积层无激活函数，适合回归任务}$ - $\text{多尺度预测结果存储在}$ `x_unwrap` $\text{中，可用于多尺度监督训练}$ - $\...

【TF-Slim使用】

图挖掘领域，新晋砖家 ☞ 未来可期，欢迎和静静一起学习交流吖

08-18

1283

转载 https://blog.csdn.net/MOU_IT/article/details/82717745 1、简介对于tensorflow.contrib这个库，tensorflow官方对它的描述是：此目录中的任何代码未经官方支持，可能会随时更改或删除。每个目录下都有指定的所有者。它旨在包含额外功能

tensorflow中slim模块api介绍

热门推荐

妙猪的专栏

08-29

6万+

最近需要使用slim模块，先把slim的介绍放在这，后续会进行整理 github：https://github.com/tensorflow/tensorflow/edit/master/tensorflow/contrib/slim/README.md TensorFlow-Slim TF-Slim is a lightweight library for defining, trai

如何用TensorFlow和TF-Slim实现图像分类与分割

优快云人工智能

11-28

6462

原文： Image Classification and Segmentation with Tensorflow and TF-Slim 作者： Daniil Pakhomov 译者： KK4SBB 审校：王艺欢迎技术投稿、约稿、给文章纠错，请发送邮件至heyc@csdn.net 本文将介绍如何用近日发布的TF-Slim工具包和预训练的模型来完成图像分类和图像分...

【Tensorflow slim】slim variables包

zj360202的专栏

11-21

2258

在本地tensorflow中创建变量需要预定义值或初始化机制（例如，从高斯随机采样）。此外，如果需要在特定设备（如GPU）上创建变量，则必须明确规定。为了减少变量创建所需的代码，TF-Slim在variables.py中提供了一组简单的包装函数，这些函数允许调用者轻松定义变量。例如，要创建权重变量，使用截断的正态分布对其进行初始化，用l2_loss调整它，并将其放在CPU上，只需声

Tensorflow使用slim工具(vgg16模型）实现图像分类与分割

weixin_36368407的博客

12-16

2万+

接触tensorflow小白，网上教程很多，图像分类应该属于比较经典的一个例子啦，特别是google推了slim，但是网上的教程遗漏啦许多细节介绍会导致，经过调试终于跑出来啦结果还算可以，分享一下我的环境，cuda8.0+cudnn5.1+python2.7 关于tensorflow,cuda+cudnn等安装推荐教程： http://blog.csdn.net/xierhacke

TensorFlow 学习笔记（二）

u013548568的博客

04-25

852

1、 slim.get_variables()，slim.get_model_variables()，tf.global_variables()，tf.trainable_variables()，tf.all_variables() [989,446,989,270,989] tf.GraphKeys.GLOBAL_VARIABLES:可以在多个设备上共享的变量 tf.GraphKeys....

【Tensorflow】辅助工具篇——tensorflow slim(TF-Slim)介绍

xf__mao的博客

06-18

5万+

可能很多tensorflow的老版本玩家没见过这个东西，slim这个模块是在16年新推出的，其主要目的是来做所谓的“代码瘦身”。但事实上它已经成为我比较喜欢，甚至是比较常用的模块，github上面大部分tensorflow的工程都会涉及到它，不得不说，撇开Keras，TensorLayer，tfLearn这些个高级库不谈，光用tensorflow能不能写出简洁的代码？当然行，有slim就够了！

学习slim（一）官方说明

jjdbear的专栏

01-17

2万+

从11月正式开始学习DL这块，由于项目这块目前都是用tensorflow写工程，所以学习DL同时也在学习tf，感觉在tf基础上出现了很多封装好的库，比如slim，主要也是同事们在用，所以自己也学习一下。详细说明介绍可以参考tensorflow官方提供的说明文档，本文基本上是对说明的翻译 https://github.com/tensorflow/tensorflow/blob/master/

TensorFlow-Slim介绍

qq_15933615的博客

03-09

1537

原文地址： https://github.com/tensorflow/tensorflow/blob/master/tensorflow/contrib/slim/README.md TensorFlow-Slim TF-Slim是一个轻量级库，用于定义，训练和评估TensorFlow中的复杂模型。 tf-slim的组件可以与本地tensorflow以及其他框架（如tf.contrib.l...