主页 / user-394660

VEL's questions

Martin Hope
VEL
Asked: 2022-05-18 21:46:32 +0800 CST
  • 1

如何将列“数据”从对象转换为 int64，这里是df数据

import numpy as np
import pandas as pd


df = pd.read_excel("df.xlsx")

# пробую так, но выдает ошибку.
df["data"].values.astype(np.int64)


Traceback (most recent call last):
  File "<ipython-input-143-763b952fb481>", line 1, in <module>
    df["data"].values.astype(np.int64)
ValueError: invalid literal for int() with base 10: '1 246'
python
Martin Hope
VEL
Asked: 2022-01-27 21:27:15 +0800 CST
  • 3

有一个 DataFrame。

如何排序和构造它？

import pandas as pd

# загрузка XLSX 
df = pd.read_excel('test.xlsx', 'Лист1', parse_dates=['Date'], index_col='Date')

df
Out[50]: 
             data  Year      month
Date                              
2018-01-31  28523  2018    January
2018-02-28  24395  2018   February
2018-03-31   2103  2018      March
2018-04-30  14450  2018      April
2018-05-31  -1903  2018        May
2018-06-30  -3112  2018       June
2018-07-31  13752  2018       July
2018-08-31   3892  2018     August
2018-09-30  26652  2018  September
2018-10-31 -12245  2018    October
2018-11-30   7437  2018   November
2018-12-31 -10296  2018   December
2019-01-31  32621  2019    January
2019-02-28   4176  2019   February
2019-03-31   7339  2019      March
2019-04-30   6582  2019      April
2019-05-31   8254  2019        May
2019-06-30  21825  2019       June
2019-07-31 -17253  2019       July
2019-08-31  -2725  2019     August
2019-09-30  35000  2019  September
2019-10-31  21760  2019    October
2019-11-30  -2458  2019   November
2019-12-31  17852  2019   December
2020-01-31  12260  2020    January
2020-02-29   2349  2020   February
2020-03-31  58246  2020      March
2020-04-30  43770  2020      April
2020-05-31  21801  2020        May
2020-06-30   7011  2020       June
2020-07-31   5521  2020       July
2020-08-31  13734  2020     August
2020-09-30   8150  2020  September
2020-10-31  -3778  2020    October
2020-11-30  39035  2020   November
2020-12-31  17447  2020   December

您需要得到以下结果：

在此处输入图像描述

python
Martin Hope
VEL
Asked: 2020-08-14 20:02:09 +0800 CST
  • 0

你能告诉我如何在图表上添加额外的点（标签）吗？

# данные
a = c(641,642,642,642,640,639,640,641,640,640,640,639,638,645,649,641,647,655,660,667,660,662,665,669,680,683,673,675,685,690)
b = c( 1, NA, NA, NA, NA, -1, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,  1, NA, NA, NA, -1, NA, NA,  1, NA, NA, NA, NA)
data = cbind(a,b)
       a  b
 [1,] 641  1
 [2,] 642 NA
 [3,] 642 NA
 [4,] 642 NA
 [5,] 640 NA
 [6,] 639 -1
 [7,] 640 NA
 [8,] 641 NA
 [9,] 640 NA
[10,] 640 NA

# строим график по данным (a)
plot(a,t="l")

如何在（b）列的图表上放置点（标记）？我附上了一个示例图表。

在此处输入图像描述

графика
Martin Hope
VEL
Asked: 2020-08-12 16:22:41 +0800 CST
  • 1

请告诉我如何替换重复项，当从 1 更改为 -1 或从 -1 更改为 1 时，将重复替换为 (NA)

mas = c(1 , 1 , 1 , 1,  1, -1, -1, -1, -1, -1, -1 ,-1 ,-1 ,-1 ,-1 ,-1, -1, -1,  1,  1,  1,  1, -1, -1)
mas
[1]  1  1  1  1  1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1  1  1  1  1 -1 -1

duplicated(mas ) # не помогает


вот что должно получиться
[1]  1 NA NA NA NA -1 NA NA NA NA NA NA NA NA NA NA NA NA  1 NA NA NA -1 NA

我尝试这种方式，它不起作用。

mas = c(1 , 1 , 1 , 1,  1, -1, -1, -1, -1, -1, -1 ,-1 ,-1 ,-1 ,-1 ,-1, -1, -1,  1,  1,  1,  1, -1, -1)
mas
[1]  1  1  1  1  1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1  1  1  1  1 -1 -1

# вот такой способ, но пропускает первое число и выводит TRUE либо FALSE, а в целом работает. 
dt = data.table(mas)

dt2 = dt$mas[-1] != dt$mas[-nrow(dt)]
dt2
[1] FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE  TRUE FALSE


а вот что должно получиться
[1]  1 NA NA NA NA -1 NA NA NA NA NA NA NA NA NA NA NA NA  1 NA NA NA -1 NA
dataframe
Martin Hope
VEL
Asked: 2020-07-21 19:17:51 +0800 CST
  • 2

我在这里加载的数据df_5.txt是链接https://yadi.sk/d/vOckRo-G0YQ2jg，但是日期和时间被分成不同的列。

#загрузка текстовых данных из файла
result <- read.table(file = "df_5.txt", row.names=NULL, fill = TRUE )

在此处输入图像描述

并且有必要在一列中。这是 ymd h:m:s 格式。在“日期”列中，获取时间序列输出。这是应该发生的事情的示例：

在此处输入图像描述

r
Martin Hope
VEL
Asked: 2020-07-08 16:33:06 +0800 CST
  • 0

如何在两张图上显示大量线条时，上下图的线条颜色是否匹配？我正在使用。

import plotly.graph_objects as go
import plotly.express as px
from plotly.subplots import make_subplots
import pandas as pd
import numpy as np
n_rows = 100
n_cols = 5
n_rows2 = 100
n_cols2 = 5


df1 = pd.DataFrame(np.random.randint(0, 100, size=(n_rows, n_cols)), columns=['col%d' % i for i in range(n_cols)])
df2 = pd.DataFrame(np.random.randint(0, 100, size=(n_rows2, n_cols2)), columns=['col%d' % i for i in range(n_cols2)])


fig_line = make_subplots(rows=2, cols=1, shared_xaxes=True, vertical_spacing=0.007, row_width=[0.23, 0.8])

# верхнии линии
for i in df1:
    fig_line.add_trace(go.Scatter(x=df1.index, y=df1[i], line=dict(width=1, dash="solid"), name = df1[i].name,  connectgaps=True), row=1, col=1) 

# нижнии линии
for n in df2:
    fig_line.add_trace(go.Scatter(x=df2.index, y=df2[n], line=dict(width=1, dash="solid" ), name = df2[n].name, connectgaps=True), row=2, col=1)

# показ в браузер
fig_line.write_html('first_figure.html', auto_open=True)
python
Martin Hope
VEL
Asked: 2020-07-07 20:40:59 +0800 CST
  • 2

如何连接 DataFrame 中的列名？

import pandas as pd
import numpy as np

n_rows = 10
n_cols = 4
df_1 = pd.DataFrame(np.random.randint(0, 100, size=(n_rows, n_cols)), columns=['col%d' % i for i in range(n_cols)])
df_2 = pd.DataFrame(np.random.randint(0, 100, size=(n_rows, n_cols)), columns=['col%d' % i for i in range(n_cols)])


def pir1(df_1, df_2):
    pidx = np.indices((df_1.shape[1], df_2.shape[1])).reshape(2, -1)
    lcol = pd.MultiIndex.from_product([df_1.columns, df_2.columns], names=[df_1.columns.name, df_2.columns.name])
    return pd.DataFrame(df_1.values[:, pidx[0]] + df_2.values[:, pidx[1]], columns=lcol)

print(pir1(df_1, df_2))

  col0                col1            ... col2           col3               
  col0 col1 col2 col3 col0 col1 col2  ... col1 col2 col3 col0 col1 col2 col3
0   99   83   64   23  139  123  104  ...  157  138   97  134  118   99   58
1   92   72   54  109   68   48   30  ...   91   73  128  151  131  113  168
2   92   73   92   91  125  106  125  ...   65   84   83   72   53   72   71
3  110  121  133  167   99  110  122  ...  129  141  175   75   86   98  132

[4 rows x 16 columns]

如何使列名加入？输出应该是这样的：

     col0_col0  col0_col1  col0_col2  ...  col3_col1  col3_col2  col3_col3
0         99         83         64    ...       118         99         58
1         92         72         54    ...       131        113        168
2         92         73         92    ...        53         72         71
3        110        121        133    ...        86         98        132
python
Martin Hope
VEL
Asked: 2020-07-01 18:56:12 +0800 CST
  • 1

有没有另一种方法来实现这个 Python3 循环？

import pandas as pd
import numpy as np
from itertools import product
from datetime import datetime

#замеряем скорость
start_time=datetime.now()

n_rows = 100
n_cols = 100
df1 = pd.DataFrame(np.random.randint(0, 100, size=(n_rows, n_cols)), columns=['col%d' % i for i in range(n_cols)])
df2 = pd.DataFrame(np.random.randint(0, 100, size=(n_rows, n_cols)), columns=['col%d' % i for i in range(n_cols)])  




#1 Есть ли возможности повысить скорость цикла for или можно предложить другую реализацию.
#2 Получившийся результат положить в другой df.(сейчас не удобно, получившийся результат добавляется в тот же df2)
for x1, x2 in itertools.product(df1, df2):
    df2['{0}_{1}'.format(x1,x2)] = df1[x1] + df2[x2] 




# замеряем скорость
end_time = datetime.now()
print('Скорость выполнения скрипта: {}'.format(end_time - start_time))

In [12]:Скорость выполнения скрипта: 0:00:06.134649
python
Martin Hope
VEL
Asked: 2020-07-01 00:30:51 +0800 CST
  • 0

告诉我如何对名称唯一的列求和？

还有df1_df2

df1 = pd.DataFrame(data={'A': [1, 2, 3, 4], 'B': [5, 6, 7, 8], 'C': [9, 10, 11, 12], 'D': [13, 14, 15, 16], 'E': [17, 18, 19, 20],  'F': [21, 22, 23, 24], 'G': [25, 26, 27, 28], 'H': [29, 30, 31, 32], 'L': [33, 34, 35, 36], 'K': [37, 38, 39, 40], 'M': [41, 42, 43, 44], 'P': [45, 46, 47, 48]})
df2 = pd.DataFrame(data={'A': [254, 223, 335,222], 'B': [545, 667, 743, 856], 'D': [986, 104, 113, 124], 'E': [137, 149, 165, 176]})

在这一步，df1我花费了列的总和，df_summ一切都很好。

my_list=[(pd.Series(df1.loc[:,list(i)].sum(axis=1), name='_'.join(df1.loc[:,list(i)].columns))) for i in list(itertools.combinations(df1.columns,2))] 
df_summ=pd.concat(my_list,axis=1)
print(df_summ)

   A_B  A_C  A_D  A_E  A_F  A_G  A_H  ...  H_P  L_K  L_M  L_P  K_M  K_P  M_P
0    6   10   14   18   22   26   30  ...   74   70   74   78   78   82   86
1    8   12   16   20   24   28   32  ...   76   72   76   80   80   84   88
2   10   14   18   22   26   30   34  ...   78   74   78   82   82   86   90
3   12   16   20   24   28   32   36  ...   80   76   80   84   84   88   92

[4 rows x 66 columns]

这里的问题是如何总结df_summ和df2，以便总结独特的列。似乎itertools.product可以应用此功能。

它应该大致如下所示：

    A_B+D  A_C+D  A_D+B  A_E+B  A_F+B  A_G+B  A_H+B  
0    992    996    559    563    567    571    575   
1    112    116    683    687    691    695    699   
2    123    127    761    765    769    773    777   
3    136    140    876    880    884    888    892
python
Martin Hope
VEL
Asked: 2020-06-21 04:29:41 +0800 CST
  • 2

如何df_name_serial_number让所有唯一值 df_name_serial_number从顶部开始？

我将举两个例子，两种情况下的数据都是一样的，第一个例子中数据是有序的，第二个例子中数据是混合的（接近真实的）。必须从中的顶部开始选择唯一值df_name_serial_number。

初始数据：

import pandas as pd
from itertools import combinations

#создаем df
df = pd.DataFrame(data={'A': [1, 2, 3, 4], 'B': [5, 6, 7, 8], 'C': [9, 10, 11, 12], 'D': [13, 14, 15, 16], 'E': [17, 18, 19, 20],  'F': [21, 22, 23, 24], 'G': [25, 26, 27, 28], 'H': [29, 30, 31, 32], 'L': [33, 34, 35, 36], 'K': [37, 38, 39, 40], 'M': [41, 42, 43, 44], 'P': [45, 46, 47, 48]})

#создаем df_name_serial_number 
cols = df.columns.to_list()
nums = list(range(len(cols)))

df_name_serial_number = pd.DataFrame(
        [[f"{cols[a]}_{cols[b]}", a, b] for a, b in list(combinations(nums, 2))], 
        columns="name serial_number1 serial_number2".split())
  1. 例子

提供此示例以供理解：

df_name_serial_number
    name  serial_number1  serial_number2

0   A_B               0               1
1   A_C               0               2
2   A_D               0               3
3   A_E               0               4
4   A_F               0               5
5   A_G               0               6
6   A_H               0               7
7   A_L               0               8
8   A_K               0               9
9   A_M               0              10
10  A_P               0              11
11  B_C               1               2
12  B_D               1               3
13  B_E               1               4
14  B_F               1               5
15  B_G               1               6
16  B_H               1               7
17  B_L               1               8
18  B_K               1               9
19  B_M               1              10
20  B_P               1              11
21  C_D               2               3
22  C_E               2               4
23  C_F               2               5
24  C_G               2               6
25  C_H               2               7
26  C_L               2               8
27  C_K               2               9
28  C_M               2              10
29  C_P               2              11
30  D_E               3               4
31  D_F               3               5
32  D_G               3               6
33  D_H               3               7
34  D_L               3               8
35  D_K               3               9
36  D_M               3              10
37  D_P               3              11
38  E_F               4               5
39  E_G               4               6
40  E_H               4               7
41  E_L               4               8
42  E_K               4               9
43  E_M               4              10
44  E_P               4              11
45  F_G               5               6
46  F_H               5               7
47  F_L               5               8
48  F_K               5               9
49  F_M               5              10
50  F_P               5              11
51  G_H               6               7
52  G_L               6               8
53  G_K               6               9
54  G_M               6              10
55  G_P               6              11
56  H_L               7               8
57  H_K               7               9
58  H_M               7              10
59  H_P               7              11
60  L_K               8               9
61  L_M               8              10
62  L_P               8              11
63  K_M               9              10
64  K_P               9              11
65  M_P              10              11

df_name_serial_number 如何做出这样的结果，让STARTING FROM THE TOP中只剩下唯一值？如果仔细观察，所有三列都有唯一值：

df_name_serial_number
   name  serial_number1  serial_number2
0  A_B               0               1
1  C_D               2               3
2  E_F               4               5
3  G_H               6               7
4  L_K               8               9
5  M_P              10              11
  1. 例子

在这种情况下，或者更确切地说，它们总是会混合在一起df_name_serial_number。

    name  serial_number1  serial_number2
0   A_K               0               9
1   A_E               0               4
2   F_P               5              11
3   A_C               0               2
4   B_D               1               3
5   A_G               0               6
6   K_P               9              11
7   A_L               0               8
8   A_B               0               1
9   A_M               0              10
10  A_P               0              11
11  B_C               1               2
12  A_F               0               5
13  B_E               1               4
14  B_F               1               5
15  B_G               1               6
16  B_H               1               7
17  B_L               1               8
18  B_K               1               9
19  B_M               1              10
20  B_P               1              11
21  C_D               2               3
22  C_E               2               4
23  C_F               2               5
24  C_G               2               6
25  C_H               2               7
26  C_L               2               8
27  C_K               2               9
28  C_M               2              10
29  C_P               2              11
30  D_E               3               4
31  D_F               3               5
32  D_G               3               6
33  D_H               3               7
34  D_L               3               8
35  D_K               3               9
36  D_M               3              10
37  D_P               3              11
38  E_F               4               5
39  E_G               4               6
40  E_H               4               7
41  E_L               4               8
42  E_K               4               9
43  E_M               4              10
44  E_P               4              11
45  F_G               5               6
46  F_H               5               7
47  F_L               5               8
48  F_K               5               9
49  F_M               5              10
50  A_D               0               3
51  G_H               6               7
52  G_L               6               8
53  G_K               6               9
54  G_M               6              10
55  G_P               6              11
56  H_L               7               8
57  H_K               7               9
58  H_M               7              10
59  H_P               7              11
60  L_K               8               9
61  L_M               8              10
62  L_P               8              11
63  K_M               9              10
64  A_H               0               7
65  M_P              10              11

如何做出这个结果？只留下独一无二的价值观STARTING FROM THE TOP在一切df_name_serial_number：

df_name_serial_number
   name  serial_number1  serial_number2
0   A_K               0               9
2   F_P               5               11
4   B_D               1               3
22  C_E               2               4
51  G_H               6               7
61  L_M               8              10
python
Martin Hope
VEL
Asked: 2020-06-19 21:53:50 +0800 CST
  • 2

为了准确和清楚地解释我需要什么，我将举一个操作序列的示例：

  1. 有一个 DataFrame（在现实生活中还有很多）：

    df = pd.DataFrame(data={'A': [1, 2, 3, 4], 'B': [5, 6, 7, 8], 'C': [9, 10, 11, 12], 'D': [13, 14, 15, 16], 'E': [17, 18, 19, 20],  'F': [21, 22, 23, 24], 'G': [25, 26, 27, 28], 'H': [29, 30, 31, 32], 'L': [33, 34, 35, 36], 'K': [37, 38, 39, 40], 'M': [41, 42, 43, 44], 'P': [45, 46, 47, 48]})
df

   A  B   C   D   E   F   G   H   L   K   M   P
0  1  5   9  13  17  21  25  29  33  37  41  45
1  2  6  10  14  18  22  26  30  34  38  42  46
2  3  7  11  15  19  23  27  31  35  39  43  47
3  4  8  12  16  20  24  28  32  36  40  44  48

  2. 我做组合itertools.combinations并得到列的总和：

    my_list=[(pd.Series(df.loc[:,list(i)].sum(axis=1),\
name='_'.join(df.loc[:,list(i)].columns))) for i in list(itertools.combinations(df.columns,2))]    
final_df=pd.concat(my_list,axis=1)
print(final_df)

     A_B  A_C  A_D  A_E  A_F  A_G  A_H  ...  H_P  L_K  L_M  L_P  K_M  K_P  M_P
 0    6   10   14   18   22   26   30  ...   74   70   74   78   78   82   86
 1    8   12   16   20   24   28   32  ...   76   72   76   80   80   84   88
 2   10   14   18   22   26   30   34  ...   78   74   78   82   82   86   90
 3   12   16   20   24   28   32   36  ...   80   76   80   84   84   88   92

  3. 这是此步骤的问题。我怎样才能得到这个，我会打电话给它，df_name_serial_number？

    a）“名称”列包含df来自步骤（2。进行组合）的列的名称。
    b）在“serial_number1”列中，这是df步骤中的序列号（1.there is a DataFrame）。
    c）在“serial_number2”列中，这是步骤中df的序列号（1.有一个DataFrame）。
    换句话说，A_B它们来自步骤#1的序列号，'A'的序列号为0，'B'的序列号为1。等等。

        name  serial_number1  serial_number2
0   A_B               0               1
1   A_C               0               2
2   A_D               0               3
3   A_E               0               4
4   A_F               0               5
5   A_G               0               6
6   A_H               0               7
7   A_L               0               8
8   A_K               0               9
9   A_M               0              10
10  A_P               0              11
11  B_C               1               2
12  B_D               1               3
13  B_E               1               4
14  B_F               1               5
15  B_G               1               6
16  B_H               1               7
17  B_L               1               8
18  B_K               1               9
19  B_M               1              10
20  B_P               1              11
21  C_D               2               3
22  C_E               2               4
23  C_F               2               5
24  C_G               2               6
25  C_H               2               7
26  C_L               2               8
27  C_K               2               9
28  C_M               2              10
29  C_P               2              11
30  D_E               3               4
31  D_F               3               5
32  D_G               3               6
33  D_H               3               7
34  D_L               3               8
35  D_K               3               9
36  D_M               3              10
37  D_P               3              11
38  E_F               4               5
39  E_G               4               6
40  E_H               4               7
41  E_L               4               8
42  E_K               4               9
43  E_M               4              10
44  E_P               4              11
45  F_G               5               6
46  F_H               5               7
47  F_L               5               8
48  F_K               5               9
49  F_M               5              10
50  F_P               5              11
51  G_H               6               7
52  G_L               6               8
53  G_K               6               9
54  G_M               6              10
55  G_P               6              11
56  H_L               7               8
57  H_K               7               9
58  H_M               7              10
59  H_P               7              11
60  L_K               8               9
61  L_M               8              10
62  L_P               8              11
63  K_M               9              10
64  K_P               9              11
65  M_P              10              11

  4. 第二个问题是如何df_name_serial_number从接收到的从顶部开始的所有重复名称（！）？让它看起来像这样：

    df_name_serial_number

    name  serial_number1  serial_number2
0   A_B               0               1
21  C_D               2               3
38  E_F               4               5
51  G_H               6               7
60  L_K               8               9
65  M_P              10              11
python