如何按组的唯一值 plot？

Question

I have this df:

      CODE   MONTH   PP  ORDER
0     000130  01    1.8    1 
1     000130  02    5.2    1 
2     000130  03    4.4    1
3     000130  04    2.3    1
4     000130  05    0.6    1
     ...  ..  ...
1555  158313  08    0.0    1
1556  158313  09    0.0    1
1557  158313  10    0.0    1
1558  158313  11    0.0    1
1559  158313  12    0.6    1
0     000130  01    1.2    2
1     000130  02    6.2    2
2     000130  03    4.8    2
3     000130  04    1.3    2
4     000130  05    1.6    2
     ...  ..  ...
1555  158313  08    2.2    2
1556  158313  09    1.1    2
1557  158313  10    0.1    2
1558  158313  11    2.0    2
1559  158313  12    2.1    2

I want to plot df['MONTH'] and df['PP'] by code. I want to plot the values of df['MONTH'] and df['PP'] with code 000130 and order 1, and the second values of df['MONTH'] and df['PP'] with code 000130 and order 2, and the third etc etc in the same graph. I want to do this by each code. So i tried this code:

for code, dfs in df.groupby('CODE'):
    fig, ax = plt.subplots(figsize=(30, 15))
    
    
    plt.plot(df.iloc[0:12]['PP'],'o--',color='black')
    plt.plot(df.iloc[12:24]['PP'],'o-',color='blue', marker='o')
    plt.plot(df.iloc[24:36]['PP'],'o-',color='green', marker='o')
    plt.plot(df.iloc[36:48]['PP'],'o-',color='yellow', marker='o')
    plt.plot(df.iloc[48:60]['PP'],'o-',color='red', marker='o')
    plt.plot(df.iloc[60:72]['PP'],'o-',color='orange', marker='o')
    plt.plot(df.iloc[72:84]['PP'],'o-',color='brown', marker='o')

But it doesn't plot correctly. Would you mind to help me?

Thanks in advance.

Answer 1

It is not clear if you want everything plotted on the same figure or each plot separately. If you want everything on a different plot, this plots for each code/order the month on the x axis an the pp on the y axis.

groups = df.groupby(['CODE','ORDER'])
for i in groups.groups:
    plt.figure()
    groups.get_group(i)[['MONTH', 'PP']].plot(x='MONTH')
    plt.ylabel('PP')
    plt.title(f'Code{i[0]} = Order {i[1]}')
plt.show()

Or, more pythonic (same result):

df.groupby(['CODE','ORDER'])['MONTH','PP'].plot(x='MONTH')
plt.show()

If you want everything in the same figure instead:

fig, ax = plt.subplots(figsize=(10,8))
for code in df.CODE.unique():
    for order in df.ORDER.unique():
        df.loc[(df.CODE==code)&(df.ORDER==order)][['MONTH','PP']].plot(x='MONTH', y='PP', ax=ax, label=str(code)+'-'+str(order))
        ax.set_ylabel('PP')
        ax.set_title('*****', size=10)
plt.show()

Output: