<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[m0_49361060的博客]]></title><description><![CDATA[]]></description><link>https://blog.youkuaiyun.com/m0_49361060</link><language>zh-cn</language><generator>https://blog.youkuaiyun.com/</generator><copyright><![CDATA[Copyright &copy; m0_49361060]]></copyright><item><title><![CDATA[Zotero使用教程]]></title><link>https://blog.youkuaiyun.com/m0_49361060/article/details/128166669</link><guid>https://blog.youkuaiyun.com/m0_49361060/article/details/128166669</guid><author>m0_49361060</author><pubDate>Sat, 03 Dec 2022 22:18:24 +0800</pubDate><description><![CDATA[zotero工具使用]]></description><category></category></item><item><title><![CDATA[windows多用户远程连接]]></title><link>https://blog.youkuaiyun.com/m0_49361060/article/details/121901538</link><guid>https://blog.youkuaiyun.com/m0_49361060/article/details/121901538</guid><author>m0_49361060</author><pubDate>Tue, 14 Dec 2021 09:40:52 +0800</pubDate><description><![CDATA[windows多用户远程连接]]></description><category></category></item><item><title><![CDATA[ROC曲线]]></title><link>https://blog.youkuaiyun.com/m0_49361060/article/details/118752957</link><guid>https://blog.youkuaiyun.com/m0_49361060/article/details/118752957</guid><author>m0_49361060</author><pubDate>Thu, 15 Jul 2021 10:33:02 +0800</pubDate><description><![CDATA[ROC曲线是一种用来度量分类模型好坏的一个标准，全名叫做Receiver Operating Characteristic。ROC曲线绘制时，首先将每个样本属于正样本的概率进行排序，再依次以每个样本属于正样本的概率作为阈值。高于这个阈值被视为正样本，低于这个阈值视为负样本。由此进行计算，得到该样本的FPR和TPR，并以此为坐标点。
TPR代表分类器预测的正类中实际正实例占所有正实例的比例。
FPR代表分类器预测的正类中实际负实例占所有负实例的比例
计算公式分别为
TPR=TPTP+FN
\\TPR = \f]]></description><category></category></item><item><title><![CDATA[从多个文件夹中读取图片 并以文件夹序号命名图片]]></title><link>https://blog.youkuaiyun.com/m0_49361060/article/details/118668107</link><guid>https://blog.youkuaiyun.com/m0_49361060/article/details/118668107</guid><author>m0_49361060</author><pubDate>Mon, 12 Jul 2021 09:39:18 +0800</pubDate><description><![CDATA[//
import skimage.io
import skimage.color
import rawpy
import imageio
import os
import glob
import matplotlib.image as mp

data_dir = 'E:\image/'  # 文件地址/名称
classes = os.listdir(data_dir)
for cls in classes:
    files = os.listdir(data_dir + cls)
    j = 0]]></description><category></category></item><item><title><![CDATA[keras+tensorboard基础用法（菜鸟教程）]]></title><link>https://blog.youkuaiyun.com/m0_49361060/article/details/117745607</link><guid>https://blog.youkuaiyun.com/m0_49361060/article/details/117745607</guid><author>m0_49361060</author><pubDate>Wed, 09 Jun 2021 16:33:51 +0800</pubDate><description><![CDATA[从pycharm中调用tensorboard
我用的keras框架 使用一个tensorboard回调函数来训练模型

from keras.callbacks import TensorBoard
#日志将被写入logs这个位置
callbacks = TensorBoard(log_dir='2d-logs',,histogram_freq=1,embeddings_freq=1)
#训练模型
model.fit(x=x_train, y=y_train, batch_size=batch_size, ]]></description><category></category></item><item><title><![CDATA[用于少样本故障诊断的半监督元学习网络（有SE模块）]]></title><link>https://blog.youkuaiyun.com/m0_49361060/article/details/115596864</link><guid>https://blog.youkuaiyun.com/m0_49361060/article/details/115596864</guid><author>m0_49361060</author><pubDate>Thu, 15 Apr 2021 22:18:37 +0800</pubDate><description><![CDATA[Semi-supervised meta-learning networks with squeeze-and-excitation attention for few-shot fault diagnosis
用于少样本故障诊断的半监督元学习网络
摘要：在工程实践中，数据尤其是标注数据的缺乏阻碍了深度学习在机械故障诊断中的广泛应用。然而，收集和标记数据通常既昂贵又耗时。针对这一问题，本文提出了一种用于少样本故障诊断的半监督元学习网络(SSMN)。SSMN由一个参数化编码器、一个非参数化原型精化过程和一个距]]></description><category></category></item><item><title><![CDATA[未知工作载荷下故障诊断的零样本学习方法（阅读笔记）]]></title><link>https://blog.youkuaiyun.com/m0_49361060/article/details/115049713</link><guid>https://blog.youkuaiyun.com/m0_49361060/article/details/115049713</guid><author>m0_49361060</author><pubDate>Sun, 21 Mar 2021 21:57:37 +0800</pubDate><description><![CDATA[阅读笔记之A zero‑shot learning method for fault diagnosis under unknown working loads
未知工作载荷下故障诊断的零样本学习方法
创新点：本文提出了一种基于压缩堆叠自动编码器的零样本学习方法。该方法仅由已知工作负荷的数据训练而成，无需先验数据即可诊断未知但相关的工作负荷故障。

...]]></description><category></category></item><item><title><![CDATA[菜鸟笔记之改进欧拉公式]]></title><link>https://blog.youkuaiyun.com/m0_49361060/article/details/113749124</link><guid>https://blog.youkuaiyun.com/m0_49361060/article/details/113749124</guid><author>m0_49361060</author><pubDate>Sun, 07 Feb 2021 23:06:30 +0800</pubDate><description><![CDATA[数值分析作业（菜鸟的自我记录）
改进欧拉公式计算初值问题
搜到别人的博客
更改公式后出现错误
1.公式要加*号 否则变量会发灰
2.错误


正确



]]></description><category></category></item><item><title><![CDATA[安装anaconda问题]]></title><link>https://blog.youkuaiyun.com/m0_49361060/article/details/108348549</link><guid>https://blog.youkuaiyun.com/m0_49361060/article/details/108348549</guid><author>m0_49361060</author><pubDate>Tue, 01 Sep 2020 19:03:35 +0800</pubDate><description><![CDATA[卸载又重装anaconda
但是下好了只有prompt
研究了一天 卸载又装了好几次还是不行
然后打开cmd 发现这个问题

就去搜解决办法https://blog.youkuaiyun.com/weixin_43096051/article/details/89950392
照着这个解决了以后 再下载anaconda就正常了

自我记录一下

...]]></description><category></category></item></channel></rss>