mtj66的博客，交流WX:SpringBreeze1104

Mac M1 12.3系统安装TensorFlow各种报错，现在的Miniforge3-MacOSX-arm64.sh 默认3.9.X。同时通过指引安装的tf和python3.8不兼容。只能用python 3.9.X进行安装了。默认安装的numpy==1.23.2。和当前的TensorFlow版本不兼容。重新安装1.22.0版本的numpy。

三层简单神经网络 In [1]:import tensorflow as tf 1.1 定义变量In [2]:w1= tf.Variable(tf.random_normal([2, 3], stddev=1, seed=1)) w2= tf.Variable(tf.random_normal([3, 1], stddev=1, seed=1)) x = tf.consta...

在运行 objection_detection 下的train进行训练的时候报错 如下:D:\software\Anaconda\envs\py3\python.exe E:/tensorflow-relevant-download/light_dection/object_detection/train.py --logtostderr --pipeline_config_path=../m...

Unable to open table file ..\data\model_ckpt: Unknown: NewRandomAccessFile failed to Create/Open:这里是因为加载模型,但是目录文件夹下并不存在保存的模型, 下载指定额模型放到data目录下,并修改checkpoint的目录为..\data\model.chpt 下载模型解压如下 :$ l...

在介绍Faster R-CNN之前，先来介绍一些前验知识，为Faster R-CNN做铺垫。一、基于Region Proposal（候选区域）的深度学习目标检测算法Region Proposal（候选区域），就是预先找出图中目标可能出现的位置，通过利用图像中的纹理、边缘、颜色等信息，保证在选取较少窗口（几千个甚至几百个）的情况下保持较高的召回率（IoU，Intersection-over-Unio...

https://blog.youkuaiyun.com/Scotthuang1989/article/details/77601458和其他的程序一样，tensorflow也有两类问题 * 功能问题：这类问题适合用tfdbg来调试，如果问题出在graph之外，那么就是一个简单的python程序了，直接上pdb,　或者打log * 性能问题：这类使用timeline, 同样如果是graph之外，那么使用cPrf...

错误1See tf.nn.softmax_cross_entropy_with_logits_v2.Traceback (most recent call last): File “D:\software\PyCharm\helpers\pydev\pydevd.py”, line 1668, in main() File “D:\software\PyCharm\...

错误2 :x and y must have the same dtype, got tf.float32 != tf.int32解决方案 :object_detection\core\losses.py 301 再devide 之前,添加 如下类型转换 self._logit_scale = tf.cast(self._logit_scale, tf.float32) # TO...

https://blog.algorithmia.com/deep-dive-into-object-detection-with-open-images-using-tensorflow/在Open Images数据集上使用TensorFlow进行Object Detection详解。Deep Dive into Object Detection with Open Im

应用一个基于Python的开源人脸识别库，face_recognition https://blog.youkuaiyun.com/hongbin_xu/article/details/76284134基于Python的开源人脸识别库：离线识别率高达99.38%https://blog.youkuaiyun.com/zchang81/article/details/76251001Android车牌识别Demo源代码说明...

'''Save and Restore a model using TensorFlow.This example is using the MNIST database of handwritten digits(http://yann.lecun.com/exdb/mnist/)Author: Aymeric DamienProject: https://github.com/aymericd...

>>> import tensorflow as tfTraceback (most recent call last):  File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/pywrap_tensorflow.py", line 58, in <module>    from tensorf...

参考：https://github.com/tflearn/tflearn/issues/964解决方法："""Tensorflow graph freezerConverts Tensorflow trained models in .pbCode adapted from:https://gist.github.com/morgangiraud/249505f

原文： How To Improve Deep Learning Performance 作者： Jason Brownlee 翻译： KK4SBB 责编：何永灿克服过拟合和提高泛化能力的20条技巧和诀窍你是如何提升深度学习模型的效果？ 这是我经常被问到的一个问题。 有时候也会换一种问法： 我该如何提高模型的准确率呢？ ……或者反过来问： 如果我的网络模

import tensorflow as tf#1. 定义变量及滑动平均类v1 = tf.Variable(0, dtype=tf.float32)step = tf.Variable(0, trainable=False)ema = tf.train.ExponentialMovingAverage(0.99, step)maintain_averages_op = ema.ap...

TensorFlow是谷歌提供的开源深度学习框架TensorFlowOnSpark: 是雅虎提供的集成到Spark上的深度学习框架鉴于我们使用的是Spark核心计算框架，现在搭建TensorFlow ON Hadoop Yarn开发环境 架构 https://www.jianshu.com/p/62b4ebb5a2f4 http://yahoohadoop.tumblr.com/

Building a Real-Time Object Recognition App with Tensorflow and OpenCVIn this article, I will walk through the steps how you can easily build your own real-time object recognition application with

1. 在上下文管理器“foo”中创建变量“v”。import tensorflow as tfIn [3]:with tf.variable_scope("foo"): v = tf.get_variable("v", [1], initializer=tf.constant_initializer(1.0))#with tf.variable_scope("foo"):...

1. 保存计算两个变量和的模型。import tensorflow as tfv1 = tf.Variable(tf.constant(1.0, shape=[1], name='v1'))v2 = tf.Variable(tf.constant(2.0, shape=[1], name='v2'))result = v1 + v2saver = tf.train.Saver(...

1. pb文件的保存方法。import tensorflow as tffrom tensorflow.python.framework import graph_utilv1 = tf.Variable(tf.constant(1.0, shape=[1]), name = "v1")v2 = tf.Variable(tf.constant(2.0, shape=[1]), na...

定义两个不同的图import tensorflow as tfg1 = tf.Graph()with g1.as_default(): v = tf.get_variable("v", [1], initializer = tf.zeros_initializer()) # 设置初始值为0g2 = tf.Graph()with g2.as_default(): ...

1. 三层简单神经网络import tensorflow as tf# 1.1 定义变量w1= tf.Variable(tf.random_normal([2, 3], stddev=1, seed=1))w2= tf.Variable(tf.random_normal([3, 1], stddev=1, seed=1))x = tf.constant([[0.7, 0.9]]) ...

根据官方文档的介绍，最简单的修改，编译并且运行演示应用的方式是通过Android Studio。虽然官方文档强调通过源码编译中涉及到的Bazel组件并不支持windows, 但是为了便于操作，除非涉及到通过源码编译，我们依然使用windows进行操作。在本文中，我们使用Android Studio打开样例工程。PREPARE:准备好梯子（没有的可以联系QQ 2549594681），An...

自定义损失函数 import tensorflow as tffrom numpy.random import RandomState# 1. 定义神经网络的相关参数和变量。¶batch_size = 8 x = tf.placeholder(tf.float32, shape=(None, 2), name="x-input")y_ = tf.placeholder(tf.f...

实现一个三层网络 #!/usr/bin/env python#-*- coding:utf-8 -*-# @author:Springimport tensorflow as tffrom numpy.random import RandomStatebatch_size = 8w1 = tf.Variable(tf.random_normal([2, 3], stddev=...

tensorflow 学习率设置 假设我们要最小化函数 y=x2y=x2 , 选择初始点 x0=5x0=5 1. 学习率为1的时候，x在5和-5之间震荡。import tensorflow as tfTRAINING_STEPS = 10LEARNING_RATE = 1x = tf.Variable(tf.constant(5, dtype=tf.float32), nam...

#!/usr/bin/env python#-*- coding:utf-8 -*-# @author:Springimport tensorflow as tfimport matplotlib.pyplot as pltimport numpy as npdata = []label = []np.random.seed(0)# 以原点为圆心，半径为1的圆把散点划分成...

搭建请参考上一篇文章。1.广播环境变量,指定Python的路径export PYTHON_ROOT=/data/Pythonexport PYSPARK_PYTHON=${PYTHON_ROOT}/bin/pythonexport SPARK_YARN_USER_ENV=”PYSPARK_PYTHON=Python/bin/python”2.提交具体任务遇到权限问题是普