python使用pandas根据子字符串对字符串列表进行排序答案

【问题标题】：python sort a list of strings based on substrings using pandaspython使用pandas根据子字符串对字符串列表进行排序
【发布时间】：2021-10-13 13:28:51
【问题描述】：

我有一个 4 列的 Excel 表，文件名、SNR、动态范围、级别。

Filename	SNR	Dynamic Range	Level
1___SLATE_FPGA_BESBEV_TX_AMIC_9.6MHz_Normal_IN1_G0_0_HQ_DEC0_FS8_HPOF.xlsx	5	11	8
19___SLATE_FPGA_BESBEV_TX_AMIC_9.6MHz_Normal_IN1_G0_0_HQ_DEC0_FS32_HPOF.xlsx	15	31	23
10___SLATE_FPGA_BESBEV_TX_AMIC_9.6MHz_Normal_IN1_G0_0_HQ_DEC0_FS16_HPOF.xlsx	10	21	24
28___SLATE_FPGA_BESBEV_TX_AMIC_9.6MHz_Normal_IN1_G0_0_HQ_DEC0_FS48_HPOF.xlsx	20	41	23
37___SLATE_FPGA_BESBEV_TX_AMIC_9.6MHz_Normal_IN1_G0_0_HQ_DEC0_FS8_HP4.xlsx	25	51	12

我需要重新组织表格的第一列 Xls 文件名，以使粗体部分按从小到大的顺序排列。即

Filename	SNR	Dynamic Range	Level
1___SLATE_FPGA_BESBEV_TX_AMIC_9.6MHz_Normal_IN1_G0_0_HQ_DEC0_FS8_HPOF.xlsx	5	11	8
37___SLATE_FPGA_BESBEV_TX_AMIC_9.6MHz_Normal_IN1_G0_0_HQ_DEC0_FS8_HP4.xlsx	25	51	12
10___SLATE_FPGA_BESBEV_TX_AMIC_9.6MHz_Normal_IN1_G0_0_HQ_DEC0_FS16_HPOF.xlsx	10	21	24
19___SLATE_FPGA_BESBEV_TX_AMIC_9.6MHz_Normal_IN1_G0_0_HQ_DEC0_FS32_HPOF.xlsx	15	31	23
28___SLATE_FPGA_BESBEV_TX_AMIC_9.6MHz_Normal_IN1_G0_0_HQ_DEC0_FS48_HPOF.xlsx	20	41	23

我不想更改实际的 excel 文件。我希望使用 pandas，因为我稍后会做一些其他操作。

我试过了

df.sort_values(by='Xls Filename', key=lambda col: col.str.contains('_FS'),ascending=True)

但是没有用。

提前谢谢你！

【问题讨论】：

标签： python excel pandas numpy

【解决方案1】：

提取模式，使用argsort找到排序索引，然后使用排序索引进行排序：

# extract the number to sort by into a Series
fs = df.Filename.str.extract('FS(\d+)_\w+\.xlsx$', expand=False)

# find the sort index using `argsort` and reorder data frame with the sort index
df.loc[fs.astype(int).argsort()]

#                                                                       Filename  ...  Level
#0    1___SLATE_FPGA_BESBEV_TX_AMIC_9.6MHz_Normal_IN1_G0_0_HQ_DEC0_FS8_HPOF.xlsx  ...      8
#4    37___SLATE_FPGA_BESBEV_TX_AMIC_9.6MHz_Normal_IN1_G0_0_HQ_DEC0_FS8_HP4.xlsx  ...     12
#2  10___SLATE_FPGA_BESBEV_TX_AMIC_9.6MHz_Normal_IN1_G0_0_HQ_DEC0_FS16_HPOF.xlsx  ...     24
#1  19___SLATE_FPGA_BESBEV_TX_AMIC_9.6MHz_Normal_IN1_G0_0_HQ_DEC0_FS32_HPOF.xlsx  ...     23
#3  28___SLATE_FPGA_BESBEV_TX_AMIC_9.6MHz_Normal_IN1_G0_0_HQ_DEC0_FS48_HPOF.xlsx  ...     23

正则表达式FS(\d+)_\w+\.xlsx$ 将捕获紧跟FS 和_\w+\.xlsx 之前的数字。

如果您可能有不匹配的模式，请转换为 float 而不是 int，因为可能有nans：

df.loc[fs.astype(float).values.argsort()]

【讨论】：

太棒了。很高兴它有帮助！