【LSM6DSOX的MLC机器学习理解】--机器学习使用教程分享
本帖最后由 justd0 于 2020-5-2 13:20 编辑<p><span style="font-size:14.0pt"><span lang="zh-CN" style="font-family:"Microsoft YaHei""><a href="https://bbs.eeworld.com.cn/thread-1120838-1-1.html" target="_blank">【LSM6DSOX的MLC机器学习理解】--机器学习简介</a>中简单介绍了机器学习模块的特点和大致的实现流程。上个帖中我简单介绍了下</span><span lang="en-US" style="font-family:"Times New Roman"">LSM6DSOX</span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">机器学习功能实现需要用</span><span lang="zh-CN" style="font-weight:bold"><span style="font-family:"Microsoft YaHei"">有监督学习方法</span></span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">,即人为采集动作数据样本,人为标注动作类别,统计出不同类别间的数据差异,构建成决策树来实现。</span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""> </span></span></p>
<p><span style="font-size:14.0pt"><span lang="zh-CN" style="font-family:"Microsoft YaHei"">本贴将结合</span><span lang="en-US" style="font-family:"Times New Roman"">U</span><span lang="en-US" style="font-family:"Microsoft YaHei"">nico</span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">介绍如何实际使用</span><span lang="en-US" style="font-family:"Microsoft YaHei"">MLC</span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">模块。</span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""> </span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei"">希望能够抛砖引玉帮到大家,由于自己也是仅仅摸索出了它的使用方法,而其功能之强大未完全掌握,在讲解有误的地方请各位指正~</span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""> </span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""> </span></span></p>
<p><span style="font-size:14.0pt"><span lang="zh-CN" style="font-family:"Microsoft YaHei"">在使用</span><span lang="en-US" style="font-family:"Times New Roman"">MLC</span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">机器学习模块之前,我们需要明确要识别的类别都是什么。这里,我用</span><span lang="zh-CN" style="font-weight:bold"><span style="font-family:"Microsoft YaHei"">静坐(</span></span><span lang="en-US" style="font-weight:bold"><span style="font-family:"Microsoft YaHei"">stationary</span></span><span lang="zh-CN" style="font-weight:bold"><span style="font-family:"Microsoft YaHei"">)</span></span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">和</span><span lang="zh-CN" style="font-weight:bold"><span style="font-family:"Microsoft YaHei"">起身(</span></span><span lang="en-US" style="font-weight:bold"><span style="font-family:"Microsoft YaHei"">standup</span></span><span lang="zh-CN" style="font-weight:bold"><span style="font-family:"Microsoft YaHei"">)</span></span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">两个动作举例。</span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""> </span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei"">有监督学习第一步就是要<strong>构建数据集</strong>。</span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""> </span></span></p>
<p><span style="font-size:14.0pt"><span lang="zh-CN" style="font-family:"Microsoft YaHei"">在</span><span lang="en-US" style="font-family:"Times New Roman"">U</span><span lang="en-US" style="font-family:"Microsoft YaHei"">nico</span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">的界面中,根据我们动作设定好所需要的</span><span lang="zh-CN" style="font-weight:bold"><span style="font-family:"Microsoft YaHei"">传感器量程</span></span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">和</span><span lang="zh-CN" style="font-weight:bold"><span style="font-family:"Microsoft YaHei"">采样频率</span></span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">。静坐和起身两个动作相对来将不需要很高的量程范围和采样频率,所以就以默认的参数来配置,如下图所示:</span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""> </span></span></p>
<p></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""> </span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei"">配置好参数后,<span lang="zh-CN" style="font-weight:bold">点击</span><span lang="en-US" style="font-weight:bold">start</span>,传感器便开始工作,这是我们点击<span lang="en-US" style="font-weight:bold">Load/Save</span>标签页。</span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""> </span></span></p>
<p></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""> </span></span></p>
<p><span style="font-size:14.0pt"><span lang="zh-CN" style="font-family:"Microsoft YaHei"">可以看到其中有个</span><span lang="en-US" style="font-weight:bold"><span style="font-family:"Times New Roman"">S</span></span><span lang="en-US" style="font-weight:bold"><span style="font-family:"Microsoft YaHei"">aveData</span></span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">的分块布局,点击</span><span lang="en-US" style="font-family:"Microsoft YaHei"">Browser</span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">按钮选择你要保存文件的路径和</span><span lang="zh-CN" style="font-weight:bold"><span style="font-family:"Microsoft YaHei"">文件名</span></span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">。</span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""> </span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei"">按照MLC的说明文档,我们<span lang="zh-CN" style="font-weight:bold">仅</span>需要<span lang="zh-CN" style="font-weight:bold">加速度计的</span><span lang="en-US" style="font-weight:bold">Acceleration</span>和<span lang="zh-CN" style="font-weight:bold">角速度计的</span><span lang="en-US" style="font-weight:bold">Angular Rate</span>数据,所以其他的选项<span lang="zh-CN" style="font-weight:bold">全都取消掉</span>,如下图所示。</span></span></p>
<p></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Times New Roman""> </span></span></p>
<p><span style="font-size:14.0pt"><span lang="zh-CN" style="font-family:"Microsoft YaHei"">配置好后呢我们就可以开始采集数据了,按照</span><span lang="en-US" style="font-family:"Times New Roman"">U</span><span lang="en-US" style="font-family:"Microsoft YaHei"">nico</span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">上</span><span lang="en-US" style="font-family:"Microsoft YaHei"">MLC</span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">的使用</span><span lang="en-US" style="font-family:"Microsoft YaHei"">.pdf</span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">中第</span><span lang="en-US" style="font-family:"Microsoft YaHei"">10</span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">页中所述:“注意:</span><span lang="zh-CN" style="font-weight:bold"><span style="font-family:"Microsoft YaHei"">正确地启动和停止</span></span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">数据日志记录是很重要的,这样才能在日志文件中获得所需的类(例如,当要记录某个移动时,必须在</span><span lang="zh-CN" style="font-weight:bold"><span style="font-family:"Microsoft YaHei"">单击【Start】之前启动该移动,在移动停止之前必须按下【Stop】</span></span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">)”</span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""> </span></span></p>
<p><span style="font-size:14.0pt"><span lang="zh-CN" style="font-family:"Microsoft YaHei"">这就很难受了,此例子中的静坐和起立两个动作都比较缓慢,所以直接按照上面的操作来,还是可以接受。但是一旦</span><span lang="zh-CN" style="font-weight:bold"><span style="font-family:"Microsoft YaHei"">动作速度很快</span></span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">,比如我即将要做的</span><span lang="zh-CN" style="font-weight:bold"><span style="font-family:"Microsoft YaHei"">羽毛球挥拍</span></span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">的动作识别,要实现上面的过程,画面可以想象(一手拿球,一手拿拍子,一手要点鼠标</span><span lang="en-US" style="font-family:"Times New Roman"">…</span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">)</span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""> </span></span></p>
<p><span style="font-size:14.0pt"><span lang="zh-CN" style="font-family:"Microsoft YaHei"">而且,要想训练的准确,需要</span><span lang="zh-CN" style="font-weight:bold"><span style="font-family:"Microsoft YaHei"">很多数据集</span></span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">,而根据测试,</span><span lang="en-US" style="font-family:"Times New Roman"">Unico</span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">的</span><span lang="en-US" style="font-family:"Times New Roman"">MLC</span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">模块需要每个数据集放在一个</span><span lang="en-US" style="font-family:"Times New Roman"">TXT</span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">文件中。</span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""> </span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei"">上述保存数据的过程,点击start和stop之后,会将数据保存在所设置的data.txt文件中,再次采集数据时,数据会叠加在这个文件中的后面,到时候我们还需要一个个分割出来。。。要么就采集一次,输入一个新的保存文件名。</span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""> </span></span></p>
<p><span style="font-size:14.0pt"><span lang="zh-CN" style="font-family:"Microsoft YaHei"">这操作兼职堪称反人类了。。我这样采集几百个数据肯定要疯了</span><span lang="en-US" style="font-family:"Times New Roman"">…</span></span></p>
<p lang="en-US"><span style="font-size:14.0pt"><span style="font-family:"Times New Roman""> </span></span></p>
<p><span style="font-size:14.0pt"><span lang="zh-CN" style="font-family:"Microsoft YaHei"">于是乎,我花了一天的时间,写了个</span><span lang="en-US" style="font-weight:bold"><span style="font-family:"Times New Roman""><span style="color:#fa0000">MLC</span></span></span><span lang="zh-CN" style="font-weight:bold"><span style="font-family:"Microsoft YaHei""><span style="color:#fa0000">数据修正工具</span></span></span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">,</span><span lang="zh-CN" style="font-family:"Microsoft YaHei""><span style="color:#fa0000">修补了这个缺陷,详细使用方法可以参考</span></span></span><a href="https://bbs.eeworld.com.cn/thread-1120853-1-1.html" target="_blank"><span style="font-size:20px;">【LSM6DSOX的MLC机器学习理解】--训练数据集修正小工具</span></a></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""> </span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei"">根据帖子里讲的,我们只需要在采集某个动作类别时,提前点击start,动作结束后stop,然后再次进行该动作,重复上述即可,最后保存的文件可以通过小工具处理成需要的格式~~。</span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""> </span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei"">有了上面的这个小工具,我们就可以很轻松的采集数据了。</span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""> </span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei"">通过多次start和stop,采集<span lang="zh-CN" style="font-weight:bold">静坐</span>数据,保存到stationary_data.txt中。多次strat和stop,采集<span lang="zh-CN" style="font-weight:bold">起身</span>数据,保存到standup_data.txt中。</span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""> </span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei"">最后可以得到这两个数据集文件,如下图所示:</span></span></p>
<p></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""> </span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei"">其中的数据格式如下图所示,点开查看大图,如果<span style="font-weight:bold">保存的类型不一致</span>,请检查上述保存选项是否<span style="font-weight:bold">只选择了红框中的两项</span>哦。</span></span></p>
<p></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""> </span></span></p>
<p><span style="font-size:14.0pt"><span lang="zh-CN" style="font-family:"Microsoft YaHei"">用</span><span lang="en-US" style="font-weight:bold"><span style="font-family:"Times New Roman""><span style="color:#fa0000">MLC</span></span></span><span lang="zh-CN" style="font-weight:bold"><span style="font-family:"Microsoft YaHei""><span style="color:#fa0000">数据修正工具</span></span></span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">可以轻松得到静坐数据集和起身数据集,如下图所示:</span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Times New Roman""> </span></span></p>
<p></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""> </span></span></p>
<p><span style="font-size:14.0pt"><span lang="zh-CN" style="font-family:"Microsoft YaHei"">然后我们就可以通过</span><span lang="en-US" style="font-family:"Times New Roman"">U</span><span lang="en-US" style="font-family:"Microsoft YaHei"">nico</span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">的</span><span lang="en-US" style="font-family:"Microsoft YaHei"">MLC</span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">模块来处理数据集了</span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""> </span></span></p>
<p><span style="font-size:14.0pt"><span lang="zh-CN" style="font-family:"Microsoft YaHei"">打开</span><span lang="en-US" style="font-family:"Times New Roman"">MLC</span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">模块,依次导入数据集文件,如下图所示</span></span></p>
<p></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei"">依次导入完所有动作类型的数据集后,点击configuration,开始生成训练文件。</span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""> </span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei"">按照需要的配置一步步的向下进行,</span></span></p>
<p></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""> </span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei"">在设置到窗口宽度参数时,我们需要根据动作的周期来决定,滤波器也还没太会用,就都先用默认值了,之后再仔细研究下。</span></span></p>
<p></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""> </span></span></p>
<p><span style="font-size:14.0pt"><span lang="zh-CN" style="font-family:"Microsoft YaHei"">如上图所示,底部需要我们选择</span><span lang="en-US" style="font-family:"Times New Roman"">F</span><span lang="en-US" style="font-family:"Microsoft YaHei"">eatures</span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">,动作的特征,即我们根据传感器的那些值来判断,实现决策树的分类,特征具体有哪些可以参考上一篇帖子</span><span lang="zh-CN" style="font-weight:bold"><span style="font-family:"Microsoft YaHei"">的简介</span></span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">。</span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""> </span></span></p>
<p><span style="font-size:14.0pt"><span lang="zh-CN" style="font-family:"Microsoft YaHei"">我又写了个</span><span lang="en-US" style="font-weight:bold"><span style="font-family:"Times New Roman""><span style="color:#fa0000">MLC</span></span></span><span lang="zh-CN" style="font-weight:bold"><span style="font-family:"Microsoft YaHei""><span style="color:#fa0000">数据显示工具</span></span></span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">,专门看修正后数据的各种参数使用,具体使用说明可以</span></span><a href="https://bbs.eeworld.com.cn/thread-1120854-1-1.html" target="_blank"><span style="font-size:20px;">【LSM6DSOX的MLC机器学习理解】--训练数据集显示小工具</span></a><span style="font-size:14.0pt"><span lang="zh-CN" style="font-family:"Microsoft YaHei""><span style="color:#fa0000">。</span></span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">想要看自己采集数据的各种特征呢,便可通过</span><span lang="en-US" style="font-weight:bold"><span style="font-family:"Times New Roman""><span style="color:#fa0000">MLC</span></span></span><span lang="zh-CN" style="font-weight:bold"><span style="font-family:"Microsoft YaHei""><span style="color:#fa0000">数据显示工具</span></span></span><span lang="zh-CN" style="font-weight:bold"><span style="font-family:"Microsoft YaHei"">显示</span></span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">所有特征值,我把能计算的量都算了出来,可以对数据有个直观的参考</span><span lang="en-US" style="font-family:"Microsoft YaHei"">~~</span></span></p>
<p></p>
<p lang="en-US"><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""> </span></span></p>
<p><span style="font-size:14.0pt"><span lang="zh-CN" style="font-family:"Microsoft YaHei"">这里呢,由于静坐和起身两个动作的角速度值具有很明显的差异,于是我就暴力的选择了所有角速度的均值、方差、能量和峰峰值,最后保存</span><span lang="en-US" style="font-family:"Times New Roman"">ARFF</span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">文件,如下图</span></span></p>
<p></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""> </span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei"">这里需要注意,在填写了arff file,需要<span lang="zh-CN" style="font-weight:bold">点击</span><span lang="en-US" style="font-weight:bold">next</span>之后才会保存文件。保存了之后,unico就暂时先不使用了,但也<span lang="zh-CN" style="font-weight:bold">别关</span>。后面需要用。</span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""> </span></span></p>
<p><span style="font-size:14.0pt"><span lang="zh-CN" style="font-family:"Microsoft YaHei"">接下来需要用到的是机器学习工具,这里我就用官方文档中推荐的</span><span lang="zh-CN" style="font-family:"Times New Roman"">Weka</span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">来举例介绍</span><span lang="en-US" style="font-family:"Microsoft YaHei"">~</span></span></p>
<p lang="en-US"><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""> </span></span></p>
<p><span style="font-size:14.0pt"><span lang="zh-CN" style="font-family:"Microsoft YaHei"">用</span><span lang="zh-CN" style="font-family:"Times New Roman"">Weka</span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">打开生成的</span><span lang="en-US" style="font-family:"Microsoft YaHei"">action.arff</span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">文件,可以看到如下界面。</span></span></p>
<p></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei"">由于静坐和起身这连个动作的上述指标差异都很大,所以可以看到class项两个类别的数值还是有一定差异的,所以我们可以直接生成决策树。如果动作复杂,则需要针对性的选择不同的轴的不同统计特征来进行判断哦。</span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""> </span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei"">点击classify界面,点击choose,在trees下选择<span lang="en-US" style="font-weight:bold">J48</span><span lang="zh-CN" style="font-weight:bold">二叉树分类</span>方法。</span></span></p>
<p></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""> </span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei"">然后点击start,即可得到学习结果了,如下图所示。</span></span></p>
<p></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Times New Roman""> </span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei"">我们将上面决策树部分信息复制到一个txt文件中,如下图</span></span></p>
<p></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""> </span></span></p>
<p><span style="font-size:14.0pt"><span lang="zh-CN" style="font-family:"Microsoft YaHei"">然后在仍打开的</span><span lang="en-US" style="font-family:"Times New Roman"">U</span><span lang="en-US" style="font-family:"Microsoft YaHei"">nico</span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">的</span><span lang="en-US" style="font-family:"Microsoft YaHei"">MLC</span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">界面下导入决策树,如下图</span></span></p>
<p></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""><span lang="en-US" style="font-weight:bold">Meta-classifier</span>是<span lang="zh-CN" style="font-weight:bold">元分类器</span>,可以对决策树的结果进行进一步滤波,我这里就默认不设置了,可以根据实际需要进行设置,于是我们就可以将机器学习的配置文件导出来了,如图中的<span lang="en-US" style="font-weight:bold">.ucf</span>文件:</span></span></p>
<p></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Times New Roman""> </span></span></p>
<p><span style="font-size:14.0pt"><span lang="zh-CN" style="font-family:"Microsoft YaHei"">最后通过</span><span lang="en-US" style="font-family:"Times New Roman"">U</span><span lang="en-US" style="font-family:"Microsoft YaHei"">nico</span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">的</span><span lang="en-US" style="font-family:"Microsoft YaHei"">load/save</span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">界面中导入配置,即可将机器学习配置导入进来啦</span><span lang="en-US" style="font-family:"Microsoft YaHei"">~</span></span></p>
<p></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Times New Roman""> </span></span></p>
<p><span style="font-size:14.0pt"><span lang="zh-CN" style="font-family:"Microsoft YaHei"">加载完成后,我们可以在</span><span lang="en-US" style="font-family:"Times New Roman"">D</span><span lang="en-US" style="font-family:"Microsoft YaHei"">ata</span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">界面看到决策树的数据结果,下图是静坐时,决策树</span><span lang="en-US" style="font-family:"Microsoft YaHei"">1 </span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">的数据为</span><span lang="en-US" style="font-weight:bold"><span style="font-family:"Microsoft YaHei"">0</span></span><span lang="en-US" style="font-family:"Microsoft YaHei"">.</span></span></p>
<p></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei"">当我带着传感器起身后,决策树输出会变成前面设置的 <span lang="en-US" style="font-weight:bold">4</span>,如下图所示</span></span></p>
<p></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Times New Roman""> </span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei"">但由于我没有做任何滤波,所以结果很容易被触发成起身的动作,增加滤波器之后应该会好很多了。</span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""> </span></span></p>
<p><span style="font-size:14.0pt"><span lang="zh-CN" style="font-family:"Microsoft YaHei"">LSM6DSOX的</span><span lang="en-US" style="font-family:"Times New Roman"">MLC</span><span lang="zh-CN" style="font-family:"Microsoft YaHei"">机器学习模块的的使用流程分享大致就这样了</span><span lang="en-US" style="font-family:"Microsoft YaHei"">~</span></span></p>
<p lang="en-US"><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei""> </span></span></p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei"">当然它还可以结合FSM有限状态机,组成包含更复杂逻辑的运动判断功能,待后面玩顺了再分享了~</span></span></p>
<p> </p>
<p><span style="font-size:14.0pt"><span style="font-family:"Microsoft YaHei"">本文中我收集的数据集已经添加为附件,里面有10个起身动作和9个静坐动作的原始数据,各位如果有兴趣可以下载试试~</span></span></p>
<p> </p>
<p><br />
</p>
<p>处理好的数据集我放到底下啦,各位回复下即可下载食用~<br />
**** Hidden Message *****<br />
</p>
<p>学习,学习。厉害了!</p>
<p>GOOD DATA</p>
<p>厉害了,我还没搞清楚MLC的具体内容, 先参考参考</p>
<p>好文章</p>
学习,学习。 <p>11</p>
<p>大佬太强了</p>
<p>1</p>
<p> </p>
<p> </p>
页:
[1]