CODEWITHSUNDEEP
pythonpandasjoinmergeconcatenation
Matthew Lynn
Matthew Lynn

Reputation: 99

Why can't I join these two data frames together?

When I try to join two dataframes, I keep getting the following problem. They are of two different types (float64 and object), but the column I am trying to join on should be the same data type for each of the two dataframes.

products_df.PROD_NBR
Out[13]: 
0        -7.358825e+10
1        -7.358821e+10
2        -7.204736e+10
3        -7.204735e+10
4        -7.204735e+10
              ...     
189047    9.940000e+22
189048    9.940000e+22
189049    9.950000e+22
189050    9.950000e+22
189051    9.950000e+22
Name: PROD_NBR, Length: 189052, dtype: float64

postransaction_df.PROD_NBR
Out[14]: 
0          1164203101
1         72047351000
2          3600025824
3          7205861079
4         82775501058
             ...     
915739     3660081331
915740    34580265065
915741    31101710042
915742     3927832300
915743    74098527503
Name: PROD_NBR, Length: 915744, dtype: object

When I try to join:

pd.merge(postransaction_df, products_df, on='PROD_NBR')
...
ValueError: You are trying to merge on object and float64 columns. If you wish to proceed you should use pd.concat

The products_df.PROD_NBR lists all product numbers for the entire company. The postransaction_df.PROD_NBR are in regards to the item sold at the time. Shouldn't I be able to join these?

I am completely stuck. Any help would be appreciated.

Upvotes: 0

Views: 65

Answers (1)

geekzeus
geekzeus

Reputation: 895

It happens when common column in both table are of different data type
you have to convert one of the datatype by using df.PROD_NBR.astype(int) or df.PROD_NBR.astype(float)

then

products_df.merge(postransaction_df, products_df, on='PROD_NBR')

Upvotes: 1

Related Questions