python - 将交互式 Jupyter Notebook 导出到 html

Question

以下代码绘制了一个交互式图形，我可以在其中打开/关闭特定行。当我在 Ipython Notebook 中工作时，这非常有效

import pandas as pd
import numpy as np
from itertools import cycle
import matplotlib.pyplot as plt, mpld3
from matplotlib.widgets import CheckButtons
import matplotlib.patches
import seaborn as sns
%matplotlib nbagg
sns.set(style="whitegrid")
df = pd.DataFrame({'freq': {0: 0.01, 1: 0.02, 2: 0.029999999999999999, 3: 0.040000000000000001, 4: 0.050000000000000003, 5: 0.059999999999999998, 6: 0.070000000000000007, 7: 0.080000000000000002, 8: 0.089999999999999997, 9: 0.10000000000000001, 10: 0.01, 11: 0.02, 12: 0.029999999999999999, 13: 0.040000000000000001, 14: 0.050000000000000003, 15: 0.059999999999999998, 16: 0.070000000000000007, 17: 0.080000000000000002, 18: 0.089999999999999997, 19: 0.10000000000000001, 20: 0.01, 21: 0.02, 22: 0.029999999999999999, 23: 0.040000000000000001, 24: 0.050000000000000003, 25: 0.059999999999999998, 26: 0.070000000000000007, 27: 0.080000000000000002, 28: 0.089999999999999997, 29: 0.10000000000000001}, 'kit': {0: 'B', 1: 'B', 2: 'B', 3: 'B', 4: 'B', 5: 'B', 6: 'B', 7: 'B', 8: 'B', 9: 'B', 10: 'A', 11: 'A', 12: 'A', 13: 'A', 14: 'A', 15: 'A', 16: 'A', 17: 'A', 18: 'A', 19: 'A', 20: 'C', 21: 'C', 22: 'C', 23: 'C', 24: 'C', 25: 'C', 26: 'C', 27: 'C', 28: 'C', 29: 'C'}, 'SNS': {0: 91.198979591799997, 1: 90.263605442199989, 2: 88.818027210899999, 3: 85.671768707499993, 4: 76.23299319729999, 5: 61.0969387755, 6: 45.1530612245, 7: 36.267006802700003, 8: 33.0782312925, 9: 30.739795918400002, 10: 90.646258503400006, 11: 90.306122449, 12: 90.178571428600009, 13: 89.498299319699996, 14: 88.435374149599994, 15: 83.588435374200003, 16: 75.212585034, 17: 60.969387755100001, 18: 47.278911564600001, 19: 37.627551020399999, 20: 90.986394557800011, 21: 90.136054421799997, 22: 89.540816326499993, 23: 88.690476190499993, 24: 86.479591836799997, 25: 82.397959183699996, 26: 73.809523809499993, 27: 63.180272108800004, 28: 50.935374149700003, 29: 41.241496598699996}, 'FPR': {0: 1.0953616823100001, 1: 0.24489252678500001, 2: 0.15106142277199999, 3: 0.104478605177, 4: 0.089172822253300005, 5: 0.079856258734300009, 6: 0.065881413455800009, 7: 0.059892194050699996, 8: 0.059892194050699996, 9: 0.0578957875824, 10: 0.94097291541899997, 11: 0.208291741532, 12: 0.14773407865800001, 13: 0.107805949291, 14: 0.093165635189999998, 15: 0.082518134025399995, 16: 0.074532508152000007, 17: 0.065881413455800009, 18: 0.062554069341799995, 19: 0.061888600519100001, 20: 0.85313103081100006, 21: 0.18899314567100001, 22: 0.14107939043000001, 23: 0.110467824582, 24: 0.099820323417899995, 25: 0.085180009316599997, 26: 0.078525321088700001, 27: 0.073201570506399985, 28: 0.071870632860800004, 29: 0.0705396952153}})

tableau20 = ["#6C6C6C", "#92D050", "#FFC000"]
tableau20 = cycle(tableau20)

kits = ["A","B", "C"]
color = iter(["#6C6C6C", "#92D050", "#FFC000"])
fig = plt.figure(figsize=(12,8))
for kit in kits:
    colour = next(color)
    for i in df.groupby('kit'):
        grouped_df = pd.DataFrame(np.array(i[1]), columns = 
                      ['freq', 'SNS', 'FPR', 'kit'])
        if grouped_df.kit.tolist()[1] == kit:
            x = [float(value) for i, value in enumerate(grouped_df.FPR)]
            y = [float(value) for i, value in enumerate(grouped_df.SNS)]
            x, y = (list(x) for x in zip(*sorted(zip(x, y))))
            label = grouped_df['kit'].tolist()[1]
            p = plt.plot(x, y, "-o",label = label, color = colour)

labels = [label.get_text() for label in plt.legend().texts]
plt.legend().set_visible(False)
for i, value in enumerate(labels):
    exec('label%s="%s"'%(i, value))

for i in range(len(labels)):
    exec('l%s=fig.axes[0].lines[i]'%(i))

rax = plt.axes([0.92, 0.7, 0.2, 0.2], frameon=False)
check = CheckButtons(rax, (labels), ('True ' * len(labels)))
for i, rec in enumerate(check.rectangles):
     rec.set_facecolor(tableau20.next())

def func(label):
    for i in range(len(labels)):
        if label == eval('label%s'%(i)): eval('l%s.set_visible(not l%s.get_visible())'%(i,i))

    plt.draw()
check.on_clicked(func)

plt.show()

问题是，我需要将笔记本导出为 html 以与对 python 一无所知的同事共享。我怎样才能将笔记本导出为 html 并让它保持交互(切换)功能(它目前丢失)？谢谢!

Answer 1

也许您不需要将 jupyter notebook 导出为 html，但可以将 notebook 链接共享给其他人，他们可以使用浏览器访问该 url。

一个 jupyter notebook 插件可以帮助你更有效地做到这一点:jupyter/dashboards ，由官方jupyter团队维护，它可以帮助你像报告一样分享你的笔记本，你可以控制显示哪个单元格以及每个单元格显示的位置。值得一试!