psycopg2 retrieves data from the database in SQL and framing it into data, when the missing bigint column becomes float.
I would like to return it to DB after processing it with Pandas, but I get an error if it is float.
ERROR:invalid input syntax for type bigint
The reason seems to be that the integer value is also .0.
I tried to read everything by string, but when I changed it from DB to data frame, .0 came with me, so I couldn't help it.
How should I deal with it?
Thank you for your cooperation.
Run Environment
Windows 10
Python 3.7
add
Data framing was avoided by setting dtype=object.
Also
table.asttype({'hoge':np.int64})
ValueError: invalid literal for int() with base 10:
It will be
I am also trying the Int64 that you mentioned in the comment.
I will add it again when I know a little more.
The missing NaN is treated as float type .
Replacing NaN with fillna and then casting with asttype
Simply replacing with asttype results in ValueError: Cannot convert non-finite values (NA or inf) to integer.
Related Home Questions
import pandas as pd
df = pd.DataFrame({'hoge': [1,2] })
print(df)
"""
hoge
1 1
2 2
"""
df = df.reindex ([0,1,2])
print(df)
"""
hoge
1 1.0
2 2.0
3NaN
"""
# df['hoge'] = df['hoge'].astype('int64')
# ValueError: Cannot convert non-finite values (NA or inf) to integer
df['hoge'] = df['hoge'].filna(0).astype('int64')# The value replaced by filna must be reworked when returning to DB.
print(df)
"""
hoge
0 1
1 2
2 0
"""
© 2024 OneMinuteCode. All rights reserved.