CODEWITHSUNDEEP
pythonpandas
Whitebeard13
Whitebeard13

Reputation: 455

Python Pandas: reindex property deletes all previously existing values

In the following pandas dataframe (df) I would like to set the index equal to the values of column "Order" and then reindex such that, in between the index values to have also the following index values (order_adj), with nans accompanying them in the columns of the dataframe:

order_adj = [30, 182]

       ID          Date            Order    Values
16672  22000118    2019-11-06          9    0.320934 
16673  22000118    2019-11-06         44    0.213926  
16674  22000118    2019-11-06         72    0.178836  
16675  22000118    2019-11-06        135    0.182151  
16676  22000118    2019-11-06        226    0.162838

I set the index of the dataframe using the following command: df.index = df.Order.values

       ID          Date            Order    Values
9      22000118    2019-11-06          9    0.320934 
44     22000118    2019-11-06         44    0.213926  
72     22000118    2019-11-06         72    0.178836  
135    22000118    2019-11-06        135    0.182151  
226    22000118    2019-11-06        226    0.162838

But when I am using df = df.reindex(index = order_adj) instead of receiving:

       ID          Date            Order    Values
9      22000118    2019-11-06          9    0.320934 
30     nan         nan               nan    nan
44     22000118    2019-11-06         44    0.213926  
72     22000118    2019-11-06         72    0.178836  
135    22000118    2019-11-06        135    0.182151
182    nan         nan               nan    nan  
226    22000118    2019-11-06        226    0.162838

I receive

       ID          Date            Order    Values
30     nan         naT               nan    nan
182    nan         naT               nan    nan

What exactly am I missing?

Upvotes: 0

Views: 386

Answers (2)

Anurag Dabas
Anurag Dabas

Reputation: 24314

You can try:

df=df.reindex([*df.index,*order_adj]).sort_index()

OR

df=df.reindex(df.index.tolist()+order_adj).sort_index()

output of df:

        ID          Date        Order   Values
9       22000118.0  2019-11-06  9.0     0.320934
30      NaN         NaN         NaN     NaN
44      22000118.0  2019-11-06  44.0    0.213926
72      22000118.0  2019-11-06  72.0    0.178836
135     22000118.0  2019-11-06  135.0   0.182151
182     NaN         NaN         NaN     NaN
226     22000118.0  2019-11-06  226.0   0.162838

Upvotes: 1

SeaBean
SeaBean

Reputation: 23217

You can use Index.union() to combine the index with order_adj in correct sequence. Then .reindex(), as follows:

idx = df.index.union(order_adj)
df = df.reindex(idx)

or do it in one step, as follows:

df = df.reindex(df.index.union(order_adj))

Result:

print(df)

             ID        Date  Order    Values
9    22000118.0  2019-11-06    9.0  0.320934
30          NaN         NaN    NaN       NaN
44   22000118.0  2019-11-06   44.0  0.213926
72   22000118.0  2019-11-06   72.0  0.178836
135  22000118.0  2019-11-06  135.0  0.182151
182         NaN         NaN    NaN       NaN
226  22000118.0  2019-11-06  226.0  0.162838

Upvotes: 0

Related Questions