Extract min, max and avg using pandas DF

Question

Here is my Dataframe printed with df.to_markdown())

type	convesion time	covert json	data to convert	process completed	time taken
INFO	0 days 00:01:00	2023-03-09 18:56:57	23000	0 days 00:02:03	0 days 00:02:09
INFO	0 days 00:01:30	2023-03-09 19:00:01	30000	0 days 00:03:28	0 days 00:03:31
INFO	0 days 00:01:30	2023-03-09 19:00:01	50000	0 days 00:03:28	0 days 00:06:31
INFO	0 days 00:00:30	2023-03-09 19:00:01	1000	0 days 00:00:28	0 days 00:00:31

I want to get

• Min data to convert its respective conversion time, process completed time and total time(conversion time +process
  completed),time taken

• Max data to convert its respective conversion time, process completed time and total time(conversion time +process completed),time taken

• Avg data to convert its respective conversion time, process completed time and total time(conversion time +process
  completed),time taken

i tried,

df.loc[df['data to convert']==data_min, 'convesion time']
df['time taken'] = pd.to_timedelta(df['time taken'])
df['time taken'].sum()

Am not able to get what am expecting.

Expected Output:

d	Data to convert	Conversion time	Process completed	total time(process+conversion)	Time taken
MIN	1000	00:00:30	00:00:28	00:00:58	00:00:31
MAX	50000	00:01:30	00:03:28	00:04:58	00:06:31
AVG	25500	00:01:00	00:01:58	00:02:58	00:03:31

Answer 1

Use:

#create default index
df = df.reset_index(drop=True)

#convert columns to timedeltas
cols = ['convesion time','process completed','time taken']
df[cols] = df[cols].apply(pd.to_timedelta)

#create new column
df['total time'] = df['convesion time'].add(df['process completed'])

#convert values to numeric
df['data to convert'] = df['data to convert'].astype(int)

#get rows with maximal and minimal values
df1 = (df.loc[df['data to convert'].agg(['idxmin','idxmax']), 
              ['data to convert'] + cols + ['total time']]
        )
df1.index = ['MIN','MAX']

#add mean row
df1.loc['AVG'] = df1.mean()
print (df1)
     data to convert  convesion time process completed      time taken  \
MIN           1000.0 0 days 00:00:30   0 days 00:00:28 0 days 00:00:31   
MAX          50000.0 0 days 00:01:30   0 days 00:03:28 0 days 00:06:31   
AVG          25500.0 0 days 00:01:00   0 days 00:01:58 0 days 00:03:31   

         total time  
MIN 0 days 00:00:58  
MAX 0 days 00:04:58  
AVG 0 days 00:02:58  

---

df2 = df1.unstack().to_frame().T
print (df2)
  data to convert                    convesion time                  \
              MIN      MAX      AVG             MIN             MAX   
0          1000.0  50000.0  26000.0 0 days 00:00:30 0 days 00:01:30   

                         process completed                  \
                     AVG               MIN             MAX   
0 0 days 00:01:07.500000   0 days 00:00:28 0 days 00:03:28   

                              time taken                  \
                     AVG             MIN             MAX   
0 0 days 00:02:21.750000 0 days 00:00:31 0 days 00:06:31   

                              total time                  \
                     AVG             MIN             MAX   
0 0 days 00:03:10.500000 0 days 00:00:58 0 days 00:04:58   

                          
                     AVG  
0 0 days 00:03:29.250000  

EDIT: If need all maximals and all minimal rows:

#create default index
df = df.reset_index(drop=True)

#convert columns to timedeltas
cols = ['convesion time','process completed','time taken']
df[cols] = df[cols].apply(pd.to_timedelta)

#create new column
df['total time'] = df['convesion time'].add(df['process completed'])

#convert values to numeric
df['data to convert'] = df['data to convert'].astype(int)

#get rows with maximal and minimal values
df1 = df[df['data to convert'].eq(df['data to convert'].max())]
df1.index = ['MAX'] * len(df1)

df2 = df[df['data to convert'].eq(df['data to convert'].min())]
df2.index = ['MIN'] * len(df2)
print (df2)

out = pd.concat([df1, df2])

#add mean row
out.loc['AVG'] = out.mean()
print (out)