[英]Pythonic way to plot data frames AND average of y values with different x values
I have the following 4 data frames which are all diferent and are in 4 files:我有以下 4 个数据框,它们都是不同的,位于 4 个文件中:
0 P 1 E 1 1
1 P 1 E 1 2
2 P 1 E 2 3
3 P 1 E 3 4
4 P 1 E 4 5
5 P 1 B 0 6
6 P 1 B 1 7
7 P 1 B 2 8
8 P 1 B 3 9
9 P 1 B 4 10
1 P 1 E 1 3
2 P 1 E 2 4
3 P 1 E 3 5
4 P 1 E 4 6
5 P 1 B 0 7
6 P 1 B 1 8
7 P 1 B 2 9
8 P 1 B 3 10
9 P 1 B 4 11
10 P 1 B 1 12
2 P 1 E 1 5
3 P 1 E 2 6
4 P 1 E 3 7
5 P 1 E 4 8
6 P 1 B 0 9
7 P 1 B 1 10
8 P 1 B 2 11
9 P 1 B 3 12
10 P 1 B 4 13
11 P 1 B 1 14
2 P 1 E 1 5
3 P 1 E 2 6
4 P 1 E 3 7
5 P 1 E 4 8
6 P 1 B 0 9
7 P 1 B 1 10
8 P 1 B 2 11
9 P 1 B 3 12
10 P 1 B 4 13
11 P 1 B 1 14
I want to plot the first and last column of each data frame (which I can do) and then plot the average of the last columns (which I can't do).我想 plot每个数据帧的第一列和最后一列(我可以做),然后 plot 最后一列的平均值(我做不到)。 Note (cf data above) that, I need the average of y values with different x values.
请注意(参见上面的数据),我需要不同 x 值的 y 值的平均值。 The second solution here , looks promising but doesn't really solve my problem because I don't want to create a
(x1,y1)
pair for each of my data frames ( I have more than 50) 这里的第二个解决方案看起来很有希望,但并没有真正解决我的问题,因为我不想为我的每个数据帧创建一个
(x1,y1)
对(我有超过 50 个)
I tried to concatenate using pd.concat
, but the name of the columns is printed in the concatenated data frame.我尝试使用
pd.concat
进行连接,但列的名称打印在连接的数据框中。
Using the solution given from the answer mentioned above使用上述答案给出的解决方案
x1 = np.arange(10)
x2 = np.arange(10)+1
x3 = np.arange(10)+2
x4 = np.arange(10)+3
y1 = x1+1
y2 = x2+2
y3 = x3+3
y4 = x4 +4
df=pd.concat([pd.Series(y1,index=x1),
pd.Series(y2,index=x2),
pd.Series(y3,index=x3),
pd.Series(y4,index=x4)], axis=1).mean(axis=1)
ax.plot(x1, y1)
ax.plot(x2, y2)
ax.plot(x3, y3)
ax.plot(x4, y4)
df.plot(color='red')
I'm looking for a graph that looks like this:我正在寻找一个看起来像这样的图表:
Felipe Lanza's solution below was given before I edited this question with the information that I need the average of y values with different x values. Felipe Lanza 在我编辑这个问题之前给出了下面的解决方案,其中包含我需要具有不同 x 值的 y 值的平均值的信息。
From what I understood your concern was dealing with different columns with the same name...据我了解,您关心的是处理具有相同名称的不同列...
I am making quite a few assumptions, but something along these lines should work:我做了很多假设,但是这些方面的东西应该可行:
(Edit) FWIW, here's a simpler version of your proposed solution: (编辑) FWIW,这是您提出的解决方案的更简单版本:
path = 'path/to/dataFrame'
all_files = glob.glob(path + "/*.csv")
cols_list = []
fig, ax = plt.subplots()
for filename in all_files:
# You can directly load only the two columns being used...
df = pd.read_csv(filename, sep=" ", usecols=[0, 5], index_col=0, names=["A", "ID"], header=None)
# ... and skip the conversion to arrays and concatenating a series
cols_list.append(df)
df.plot(ax=ax, style='o-')
cols_list_df = pd.concat(cols_list, axis=1)
cols_list_df.mean(axis=1).plot(ax=ax, style='o-')
Using Felipe Lanza's answer above and DYZ's answer here I got a solution to my problem:使用上面的 Felipe Lanza 的回答和 DYZ 的回答,我得到了解决问题的方法:
path = 'path/to/dataFrame'
all_files = glob.glob(path + "/*.csv")
cols_list = []
fig, axes = plt.subplots()
for i, filename in enumerate(all_files):
df = pd.read_csv(filename, sep=" ", names = ["A", "E", "C","O","M","ID"],header=None )
x=df["A"]
y=df["ID"]
xarray=np.array(x)
yarray=np.array(y)
df2=pd.concat([pd.Series(yarray,index=xarray)],axis=1).mean(axis=1)
cols_list.append(df2)
axes.plot(x, y,'o-')
cols_list_df=pd.concat(cols_list,axis=1)
cols_list_df.mean(axis=1).plot(style='o-')
With the dataframes posted above, the plot looks like this:使用上面发布的数据帧,plot 看起来像这样:
I'm sure there should be a smarter solution but this is good enough for me.我确信应该有一个更聪明的解决方案,但这对我来说已经足够了。
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.