python - 为多类别 x 轴条形图生成跟踪的循环代码是什么？
问题描述
我花了几天时间尝试使用 python 和 jupyter 笔记本绘制以下数据的条形图：
site end_date weight diversion_weight count
0 All Sites 2019-11-30 19609.32 15387.52 977.0
1 All Sites 2019-12-31 13188.50 10283.40 658.0
2 All Sites 2020-01-31 21975.01 17502.41 1124.0
3 All Sites 2020-02-29 1933.06 1535.96 111.0
4 Site 1 2019-11-30 8351.48 5864.38 491.0
5 Site 1 2019-12-31 6746.97 4761.47 393.0
6 Site 1 2020-01-31 12040.58 8748.83 745.0
7 Site 1 2020-02-29 1193.33 900.73 72.0
8 Site 2 2019-11-30 11257.84 9523.14 486.0
9 Site 2 2019-12-31 6441.53 5521.93 265.0
10 Site 2 2020-01-31 9934.43 8753.58 379.0
11 Site 2 2020-02-29 739.73 635.23 39.0
对于 x 轴上的每个时间序列，每个站点都具有以下内容： diversion_weight 的条形图覆盖权重条形图
我宁愿没有一系列的子图。
我将这篇文章视为潜在的指南，但无法弄清楚如何使代码适应我的问题。
数据字典：
from pandas import timestamp
dfplot = {'site': {0: 'All Sites',
1: 'All Sites',
2: 'All Sites',
3: 'All Sites',
4: 'Site 1',
5: 'Site 1',
6: 'Site 1',
7: 'Site 1',
8: 'Site 2',
9: 'Site 2',
10: 'Site 2',
11: 'Site 2'},
'end_date': {0: Timestamp('2019-11-30 00:00:00'),
1: Timestamp('2019-12-31 00:00:00'),
2: Timestamp('2020-01-31 00:00:00'),
3: Timestamp('2020-02-29 00:00:00'),
4: Timestamp('2019-11-30 00:00:00'),
5: Timestamp('2019-12-31 00:00:00'),
6: Timestamp('2020-01-31 00:00:00'),
7: Timestamp('2020-02-29 00:00:00'),
8: Timestamp('2019-11-30 00:00:00'),
9: Timestamp('2019-12-31 00:00:00'),
10: Timestamp('2020-01-31 00:00:00'),
11: Timestamp('2020-02-29 00:00:00')},
'weight': {0: 19609.32,
1: 13188.5,
2: 21975.010000000002,
3: 1933.06,
4: 8351.48,
5: 6746.97,
6: 12040.58,
7: 1193.33,
8: 11257.84,
9: 6441.53,
10: 9934.43,
11: 739.73},
'diversion_weight': {0: 15387.52,
1: 10283.400000000001,
2: 17502.41,
3: 1535.96,
4: 5864.38,
5: 4761.47,
6: 8748.83,
7: 900.73,
8: 9523.14,
9: 5521.93,
10: 8753.58,
11: 635.23},
'count': {0: 977.0,
1: 658.0,
2: 1124.0,
3: 111.0,
4: 491.0,
5: 393.0,
6: 745.0,
7: 72.0,
8: 486.0,
9: 265.0,
10: 379.0,
11: 39.0}}
dfplot