kaggle入门之data vislization

这篇博客介绍了kaggle入门级别的数据可视化教程,涵盖了直方图、线型图、双变量图的绘制,深入讨论了如何调整图表、使用seaborn库以及处理时间序列数据。还探讨了开源绘图库poltly和复杂变量的可视化技巧。
摘要由CSDN通过智能技术生成

直方图的绘制

# 直接绘制
reviews['province'].value_counts().head(10).plot.bar()
# 百分比图的绘制
(reviews['province'].value_counts().head(10) / len(reviews)).plot.bar()
# 一种与上述方法横坐标不同的直方图
reviews['points'].value_counts().sort_index().plot.bar()
# 统计直方图(横坐标是区间不再是单一值,为防止数据偏重,故加以小于200的限制)
reviews[reviews['price'] < 200]['price'].plot.hist()

线型图的绘制

# 普通绘制
reviews['points'].value_counts().sort_index().plot.line()
# 面积图(带有阴影的折线图)
reviews['points'].value_counts().sort_index().plot.area()

双变量的绘制

# 散点图的绘制
reviews[reviews['price'] < 100].sample(100).plot.scatter(x='price', y='points')
# 六角螺栓图(螺栓颜色的深度代表重复的次数)
reviews[reviews['price'] < 100].plot.hexbin(x='price', y='points', gridsize=15)
# 多变量,直方堆叠图
wine_counts.plot.bar(stacked=True)
# 折线堆叠图(与wine_counts有关)
wine_counts.plot.area()
# 多折线图
wine_counts.plot.line()

图的调整

# 图的尺寸调整(宽度,高度)
reviews['points'].value_counts().sort_index().plot.bar(figsize=(12, 6))
# 图的颜色调整
reviews['points'].value_counts().sort_index().plot.bar(figsize=(12, 6),color='mediumvioletred')
# 图的坐标尺寸调整
reviews['points'].value_counts().sort_index().plot.bar(figsize=(12, 6),color='mediumvioletred', fontsize=16)
# 图名
reviews['points'].value_counts().sort_index().plot.bar( figsize=(12, 6),color='mediumvioletred', fontsize=16, title='Rankings Given by Wine Magazine')
# 利用matplotlib.pyplot调整图名尺寸
ax.set_title("Rankings Given by Wine Magazine", fontsize=
--------------------------- Qt Data Visualization 5.7.0 --------------------------- Qt Data Visualization module provides multiple graph types to visualize data in 3D space both with C++ and Qt Quick 2. System Requirements =================== - Qt 5.2.1 or newer - OpenGL 2.1 or newer (recommended) or OpenGL ES2 (reduced feature set) - Manipulating Qt Data Visualization graphs with QML Designer requires Qt Creator 3.3 or newer Building ======== Configure the project with qmake: qmake After running qmake, build the project with make: (Linux) make (Windows with MinGw) mingw32-make (Windows with Visual Studio) nmake (OS X) make The above generates the default makefiles for your configuration, which is typically the release build if you are using precompiled binary Qt distribution. To build both debug and release, or one specifically, use one of the following qmake lines instead. For debug builds: qmake CONFIG+=debug make or qmake CONFIG+=debug_and_release make debug For release builds: qmake CONFIG+=release make or qmake CONFIG+=debug_and_release make release For both builds (Windows/OS X only): qmake CONFIG+="debug_and_release build_all" make After building, install the module to your Qt directory: make install If you want to uninstall the module: make uninstall Building as a statically linked library ======================================= The same as above applies, you will just have to add static to the CONFIG: qmake CONFIG+=static Documentation ============= The documentation can be generated with: make docs The documentation is generated into the doc folder under the build folder. Both Qt Assistant (qtdatavisualization.qch) and in HTML format (qtdatavisualization subfolder) documentation is generated. Please refer to the generated documentation for more information: doc/qtdatavisualization/qtdatavisualization-index.html Known Issues ============ - Some platforms like Android and WinRT cannot handle multiple native windows properly, so only the Qt Quick 2 versions of graphs are available in practice for those platforms. - Shadows are not supported with OpenGL ES2 (including Angle builds in Windows). - Anti-aliasing doesn't work with OpenGL ES2 (including Angle builds in Windows). - QCustom3DVolume items are not supported with OpenGL ES2 (including Angle builds in Windows). - Surfaces with non-straight rows and columns do not always render properly. - Q3DLight class (and Light3D QML item) are currently not usable for anything. - Changing most of Q3DScene properties affecting subviewports currently has no effect. - Widget based examples layout incorrectly in iOS. - Reparenting a graph to an item in another QQuickWindow is not supported. - Android builds of QML applications importing QtDataVisualization also require "QT += datavisualization" in the pro file. This is because Qt Data Visualization QML plugin has a dependency to Qt Data Visualization C++ library, which Qt Creator doesn't automatically add to the deployment package. - Only OpenGL ES2 emulation is available for software renderer (that is, when using QCoreApplication::setAttribute(Qt::AA_UseSoftwareOpenGL))
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值