Skip to content

Instantly share code, notes, and snippets.

@mohanson

mohanson/matplotlib.ipynb

Last active May 15, 2022 02:18
Show Gist options

  • Select an option

    Learn more about clone URLs
  • Save mohanson/661b22c18e681a508f07ba04e5ce428a to your computer and use it in GitHub Desktop.

Select an option

Learn more about clone URLs
Save mohanson/661b22c18e681a508f07ba04e5ce428a to your computer and use it in GitHub Desktop.
Download ZIP
matplotlib.ipynb
Display the source blob
Display the rendered blob
Raw
matplotlib.ipynb
{
"metadata": {
"kernelspec": {
"name": "python",
"display_name": "Pyolite",
"language": "python"
},
"language_info": {
"codemirror_mode": {
"name": "python",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.8"
}
},
"nbformat_minor": 5,
"nbformat": 4,
"cells": [
{
"cell_type": "markdown",
"source": "# 如何使用 Matplotlib 绘制令人赏心悦目的图表",
"metadata": {},
"id": "2622776b-dd92-4b76-a9a7-f8e0752c1fc8"
},
{
"cell_type": "markdown",
"source": "## 从美国 GDP 数据开始\n\n有一天我在我的知乎上刷到了这个问题: [2022 年美国第一季度经济折合年率萎缩 1.4%, 还有哪些信息值得关注? 或受哪些因素导致?](https://www.zhihu.com/question/530511433)\n\n问题的信息源是**财联社**:\n\n> 美国 2022 年第一季度经济折合年率萎缩 1.4%, 预估为增长 1.1%, 前值为 6.9%.\n\n上面这句话似乎给了很多信息, 1.4%, 1.1%, 6.9% ... 但你仔细想想其实它什么都没说, 作为一个经济门外汉几乎无法从中得出对自己有意义的信息.\n\n直到有位答主贴了一张图:\n\n![img](https://pic1.zhimg.com/80/v2-2ddedb6e4c16178b0a9e52b2308eeb3b_720w.jpg)\n\n通过这张图, 读者几乎可以在一秒之内就了解近三年美国的 GDP 变化.",
"metadata": {},
"id": "790a420f-3515-4511-9ba8-112750d1da89"
},
{
"cell_type": "markdown",
"source": "## 一图胜千言\n\n\"A picture is worth a thousand words.\"\n\n许多重要人物都发表过类似的观点, 例如拿破仑曾经说过, \"A good sketch is better than a long speech\". 同时 1913 年在俄亥俄州 Piqua 的 Piqua Auto Supply House 的一份报纸广告中, 出现了类似理念:\n\n![img](https://pic3.zhimg.com/80/v2-6d8f6ccd448fd012858fd6d42d2e4a9c_720w.png)\n\n- 我们在视觉上学习和记忆.\n- 视觉智商的增长速度快于其他形式的智商.\n- 带有图片的文字影响更大. 在研究演示文稿的类型时, 明尼苏达大学和 3M 的研究人员发现, 包含\"良好\"图形的演示文稿的说服力比纯文本演示文稿高 43%.\n- 我们没有时间阅读. 尼尔森·诺曼小组研究了用户在网络上的阅读方式, 发现用户仅阅读页面内容的 25%.\n- 图表可以在更短的时间内传达更多信息.\n- 视觉信息突破了语言障碍.\n- 图表有助于设计和规划. 尤其是在研究经济问题, 或者对链上数据进行分析时.\n- 图表适合所有人.",
"metadata": {},
"id": "5cd1f242-b8ca-4736-886c-7aff650c862b"
},
{
"cell_type": "markdown",
"source": "## Matplotlib 安利\n\nMatplotlib 是一个 Python 绘图库. 无论是技术人员还是非技术人员, 都可以在短时间内快速上手.\n\nPython 有一些云端执行环境, 这意味着你可以完全不安装任何软件就能体验到 Matplotlib 的强大! [https://jupyter.org/try-jupyter/lab/](https://jupyter.org/try-jupyter/lab/)",
"metadata": {},
"id": "2883c9d6-e853-47e8-9597-7c7f5c5fa5f2"
},
{
"cell_type": "markdown",
"source": "## 折线图",
"metadata": {
"tags": []
},
"id": "800be9d5-f8bd-4765-850b-2c69a4653726"
},
{
"cell_type": "markdown",
"source": "**绘制折线图**",
"metadata": {},
"id": "5058e57c-6cf3-48c7-806b-4032877a2c3f"
},
{
"cell_type": "code",
"source": "import matplotlib.pyplot as plt\n\nx = [1, 2, 3, 4]\ny = [2, 4, 6, 8]\n\nplt.plot(x, y)\nplt.show()",
"metadata": {},
"execution_count": null,
"outputs": [],
"id": "5a44dd64-3f49-40e6-86f3-8838fe0d7552"
},
{
"cell_type": "markdown",
"source": "**皮肤**\n\nMatplotlib 默认的皮肤比较寡淡, 我一般喜欢使用 seaborn 皮肤.",
"metadata": {},
"id": "4c03bc23-9f4f-4eb3-a7bf-e99bcdecb0fc"
},
{
"cell_type": "code",
"source": "import numpy as np\nimport matplotlib.pyplot as plt\nplt.style.use('seaborn')\n\nx = np.linspace(-np.pi, np.pi, 256)\ny = np.sin(x)\n\nplt.plot(x, y)\nplt.show()",
"metadata": {},
"execution_count": null,
"outputs": [],
"id": "4bab5ae8-7861-4e73-a97d-e140f1b5fadb"
},
{
"cell_type": "markdown",
"source": "**折线样式**",
"metadata": {},
"id": "ef47b0cf-609f-4a75-95a1-5efa4b03d8a8"
},
{
"cell_type": "code",
"source": "import numpy as np\nimport matplotlib.pyplot as plt\nplt.style.use('seaborn')\n\nx = np.linspace(-np.pi, np.pi, 256)\ny = np.sin(x)\n\nplt.plot(x, y, color='pink', linewidth=2, linestyle='--')\nplt.show()",
"metadata": {},
"execution_count": null,
"outputs": [],
"id": "dfb03734-1b0a-46e4-a810-ac7ef510577b"
},
{
"cell_type": "markdown",
"source": "**填充**\n\n需要对数据进行积分的时候采用. 例如我们要去统计过去 12 个月内每个月 CKB 的 DAO 内的新增存款.",
"metadata": {},
"id": "a9dec55a-46ab-41fd-9661-4cf9d7b491a6"
},
{
"cell_type": "code",
"source": "import numpy as np\nimport matplotlib.pyplot as plt\nplt.style.use('seaborn')\n\nx = np.linspace(-np.pi, np.pi, 256)\ny = np.sin(x)\n# 参数以 .plot 相似, 区别是会填充曲线的面积, 填充分界线为 y=c(c 为 x = 0 时的数)\n# alpha 为透明度\nplt.fill(x, np.sin(x), alpha=0.5)\nplt.show()",
"metadata": {},
"execution_count": null,
"outputs": [],
"id": "53dc27e9-c9d9-4ca0-8c4e-bdaa8b6e3ff2"
},
{
"cell_type": "markdown",
"source": "**坐标位置与坐标样式**\n\n处理数学问题的时候偶尔会用到, 比如绘制椭圆曲线.",
"metadata": {},
"id": "962fa258-3a2c-484b-8df9-87ccf6c56cc7"
},
{
"cell_type": "code",
"source": "import numpy as np\nimport matplotlib.pyplot as plt\nplt.style.use('seaborn')\n\nx = np.linspace(-np.pi, np.pi, 256)\ny = np.sin(x)\nax = plt.subplot()\nax.plot(x, y)\n\n# 移动坐标轴与设置坐标轴样式\nax.spines['bottom'].set_color('#646882')\nax.spines['bottom'].set_linewidth(1)\nax.spines['bottom'].set_position(('data', 0))\nax.spines['left'].set_color('#646882')\nax.spines['left'].set_linewidth(1)\nax.spines['left'].set_position(('data', 0))\n\nplt.show()",
"metadata": {},
"execution_count": null,
"outputs": [],
"id": "8287198f-422a-43c6-a668-17fcabea662a"
},
{
"cell_type": "markdown",
"source": "**添加函数名称**",
"metadata": {},
"id": "9a7d00d0-625a-42d7-9b98-6cea47731865"
},
{
"cell_type": "code",
"source": "import numpy as np\nimport matplotlib.pyplot as plt\nplt.style.use('seaborn')\n\nx = np.linspace(-np.pi, np.pi, 256)\ny = np.sin(x)\n\n# 为曲线添加名称\nplt.plot(x, y, label='sin(x)')\nplt.legend(loc='lower right')\nplt.show()",
"metadata": {},
"execution_count": null,
"outputs": [],
"id": "a3299abf-27b4-4dfc-96a9-a36374d9f9c4"
},
{
"cell_type": "markdown",
"source": "## 散点图\n\n主要用于处理离散数据.",
"metadata": {},
"id": "88c6e8ae-d687-42f2-bd82-6917c3842716"
},
{
"cell_type": "code",
"source": "import numpy as np\nimport matplotlib.pyplot as plt\nplt.style.use('seaborn')\n\nx = np.linspace(-np.pi, np.pi, 16)\ny = np.sin(x)\n\n# s: 散点大小, 默认 20\n# c: 颜色\n# alpha: 透明度\nplt.scatter(x, y, s=50, c='#FF0000', alpha=0.5)\nplt.show()",
"metadata": {},
"execution_count": null,
"outputs": [],
"id": "264c553a-9387-4b7f-838c-55ff40af0565"
},
{
"cell_type": "markdown",
"source": "**样式**",
"metadata": {},
"id": "faf9977f-3e47-4841-a142-28c6dead9043"
},
{
"cell_type": "code",
"source": "import numpy as np\nimport matplotlib.pyplot as plt\nplt.style.use('seaborn')\n\nx = np.linspace(-np.pi, np.pi, 16)\ny = np.sin(x)\n\nplt.scatter(x, y, s=50, c='#FF0000', marker='+', alpha=0.5)\nplt.show()",
"metadata": {},
"execution_count": null,
"outputs": [],
"id": "614dd17f-3933-4ec8-8809-20ec6ebfc25e"
},
{
"cell_type": "markdown",
"source": "**三维坐标**",
"metadata": {},
"id": "9dd53fcf-e1c3-471d-b572-dd54825b5e08"
},
{
"cell_type": "code",
"source": "import matplotlib.pyplot as plt\nfrom mpl_toolkits.mplot3d import Axes3D\nimport numpy as np\nplt.style.use('seaborn')\n\np = plt.subplot(projection='3d')\n\nx = np.linspace(-np.pi, np.pi, 16)\ny = np.sin(x)\nz = np.linspace(-np.pi, np.pi, 16)\n\np.scatter(x, y, z, s=50, c='#FF0000', alpha=0.5)\np.set_zlabel('Z')\np.set_ylabel('Y')\np.set_xlabel('X')\nplt.show()",
"metadata": {},
"execution_count": null,
"outputs": [],
"id": "57ec20dc-7fb6-4710-b670-e588b97330f4"
},
{
"cell_type": "markdown",
"source": "## 柱状图",
"metadata": {},
"id": "47e42544-995b-44e2-b907-dcf29f3384ba"
},
{
"cell_type": "code",
"source": "import numpy as np\nimport matplotlib.pyplot as plt\nplt.style.use('seaborn')\n\nX = np.arange(5) + 1\nY = np.array([0.5, 0.67, 0.71, 0.56, 0.8])\n\n\nplt.bar(X, Y, tick_label=['I', 'II', 'III', 'IV', 'V'])\n# 在柱状图上标记 y 轴大小\nfor x, y in zip(X, Y):\n plt.text(x, y, f'{y:.2}', ha='center', va='bottom')\n\nplt.show()",
"metadata": {},
"execution_count": null,
"outputs": [],
"id": "d63514f2-b4fb-4a83-97cf-c51d20b90003"
},
{
"cell_type": "markdown",
"source": "**柱状图柱子的颜色**\n\nplt.bar 函数的 color 参数可以设置颜色; color 可以接受一个颜色, 也可以接受一个颜色数组",
"metadata": {},
"id": "1e9d5432-04e8-40db-b412-d60380ab19f4"
},
{
"cell_type": "code",
"source": "import numpy as np\nimport matplotlib.pyplot as plt\nplt.style.use('seaborn')\n\nX = np.arange(5) + 1\nY = np.array([0.5, 0.67, 0.71, 0.56, 0.8])\n\n\nplt.bar(X, Y, tick_label=['I', 'II', 'III', 'IV', 'V'], color=['pink', 'purple'])\n# 在柱状图上标记 y 轴大小\nfor x, y in zip(X, Y):\n plt.text(x, y, f'{y:.2}', ha='center', va='bottom')\n\nplt.show()",
"metadata": {},
"execution_count": null,
"outputs": [],
"id": "ecfb7f83-6624-4266-9672-729dafffbede"
},
{
"cell_type": "markdown",
"source": "**填充**\n\nplt.bar 函数的 hatch 参数可以填充样式, 可取值为: `/`, `\\`, `|`, `-`, `+`, `x`, `o`, `O`, `.`, `*`",
"metadata": {},
"id": "d21a9921-46d2-4a64-874f-e835446305c5"
},
{
"cell_type": "code",
"source": "import numpy as np\nimport matplotlib.pyplot as plt\nplt.style.use('seaborn')\n\nX = np.arange(5) + 1\nY = np.array([0.5, 0.67, 0.71, 0.56, 0.8])\n\n\nplt.bar(X, Y, tick_label=['I', 'II', 'III', 'IV', 'V'], hatch='/')\n# 在柱状图上标记 y 轴大小\nfor x, y in zip(X, Y):\n plt.text(x, y, f'{y:.2}', ha='center', va='bottom')\n\nplt.show()",
"metadata": {},
"execution_count": null,
"outputs": [],
"id": "71931bd1-75b5-4cac-bd6d-dea9e0c5766d"
},
{
"cell_type": "markdown",
"source": "**柱状图堆叠**\n\n使用 bottom 参数堆叠柱状图",
"metadata": {},
"id": "ab6d5441-eef6-437f-ad75-021890c1d0e8"
},
{
"cell_type": "code",
"source": "import numpy as np\nimport matplotlib.pyplot as plt\nplt.style.use('seaborn')\n\nX = np.arange(5) + 1\nY1 = np.array([0.5, 0.67, 0.71, 0.56, 0.8])\nY2 = np.random.random(5)\n\nplt.bar(X, Y1, tick_label=['I', 'II', 'III', 'IV', 'V'], label='Y1')\nplt.bar(X, Y2, bottom=Y1, label='Y2')\n\nplt.legend()\nplt.show()",
"metadata": {},
"execution_count": null,
"outputs": [],
"id": "60be7e11-67e5-4b30-ad78-e5fa530707cf"
},
{
"cell_type": "markdown",
"source": "**柱状图并列**\n\n设置柱状图的 bar_width 实现柱状图并列",
"metadata": {},
"id": "a15a689f-1eae-42bc-8336-92e33f759ada"
},
{
"cell_type": "code",
"source": "import numpy as np\nimport matplotlib.pyplot as plt\nplt.style.use('seaborn')\n\nbar_width = 0.8 / 2\n\nX1 = np.arange(5) + 1\nY1 = np.random.random(5)\nX2 = X1 + bar_width\nY2 = np.random.random(5)\n\nplt.bar(X1, Y1, bar_width)\nplt.bar(X2, Y2, bar_width)\nplt.xticks(X1+bar_width / 2, ['I', 'II', 'III', 'IV', 'V'])\nplt.show()",
"metadata": {},
"execution_count": null,
"outputs": [],
"id": "f1f181f6-07ed-4b53-a053-b337b442ef03"
},
{
"cell_type": "markdown",
"source": "**条状图**\n\n条状图与柱状图基本类似. 经常用于对候选人进行计票.",
"metadata": {},
"id": "5bb71331-a4f7-4a18-b626-c1f6e63ad0d3"
},
{
"cell_type": "code",
"source": "import numpy as np\nimport matplotlib.pyplot as plt\nplt.style.use('seaborn')\n\nX = np.arange(5) + 1\nY = np.random.random(5)\n\np = plt.subplot()\np.barh(X, Y)\np.set_yticks(X)\np.set_yticklabels(['I', 'II', 'III', 'IV', 'V'])\n\nplt.show()",
"metadata": {},
"execution_count": null,
"outputs": [],
"id": "122ed443-911e-4053-8db5-69cec53cbb4e"
},
{
"cell_type": "markdown",
"source": "## 饼图\n\n**饼图**",
"metadata": {},
"id": "5cbd6535-a437-4598-96aa-c376cdbbfc1b"
},
{
"cell_type": "code",
"source": "import matplotlib.pyplot as plt\nplt.style.use('seaborn')\n\nX = [15, 30, 45, 10]\nlabels = 'I', 'II', 'III', 'IV'\n\nplt.pie(X, labels=labels, autopct='%1.1f%%', startangle=90)\nplt.show()",
"metadata": {},
"execution_count": null,
"outputs": [],
"id": "4c2f09d0-d9c1-45b6-bc4d-6867287b3ec4"
},
{
"cell_type": "markdown",
"source": "**饼图部分强调**",
"metadata": {},
"id": "8b6d08f0-ff0c-466b-bc0c-d8bc8b279284"
},
{
"cell_type": "code",
"source": "import matplotlib.pyplot as plt\nplt.style.use('seaborn')\n\nX = [15, 30, 45, 10]\nlabels = 'I', 'II', 'III', 'IV'\nexplode = [0, 0.1, 0, 0]\n\n# explode 参数可以强调数据\nplt.pie(X, explode=explode, labels=labels, autopct='%1.1f%%', startangle=90)\nplt.axis('equal')\nplt.show()",
"metadata": {},
"execution_count": null,
"outputs": [],
"id": "13df9f0e-b7b6-43c7-a6bd-c0fe4ee2980e"
},
{
"cell_type": "markdown",
"source": "## 配色\n\n我们可以做一些更好的事情来保护读者的眼睛, 比如用比较齁人的马卡龙色\n\n![img](https://p1-tt.byteimg.com/origin/pgc-image/fbfbb2b4e9c54d379b7d7f6764781bb1.jpg)\n\n- <code style=\"color: #19CAAD; background-color: #19CAAD\">#19CAAD</code> `#19CAAD`\n- <code style=\"color: #8CC7B5; background-color: #8CC7B5\">#8CC7B5</code> `#8CC7B5`\n- <code style=\"color: #A0EEE1; background-color: #A0EEE1\">#A0EEE1</code> `#A0EEE1`\n- <code style=\"color: #BEE7E9; background-color: #BEE7E9\">#BEE7E9</code> `#BEE7E9`\n- <code style=\"color: #BEEDC7; background-color: #BEEDC7\">#BEEDC7</code> `#BEEDC7`\n- <code style=\"color: #D6D5B7; background-color: #D6D5B7\">#D6D5B7</code> `#D6D5B7`\n- <code style=\"color: #D1BA74; background-color: #D1BA74\">#D1BA74</code> `#D1BA74`\n- <code style=\"color: #E6CEAC; background-color: #E6CEAC\">#E6CEAC</code> `#E6CEAC`\n- <code style=\"color: #ECAD9E; background-color: #ECAD9E\">#ECAD9E</code> `#ECAD9E`\n- <code style=\"color: #F4606C; background-color: #F4606C\">#F4606C</code> `#F4606C`\n",
"metadata": {},
"id": "110ac700-66f5-40c3-958f-54ecc5262091"
},
{
"cell_type": "code",
"source": "",
"metadata": {},
"execution_count": null,
"outputs": [],
"id": "8d7486c8-d19b-44d4-beab-d1f3547d88e1"
}
]
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment