[Python] Pandas Overflow error converting date to month/year

I'm doing the following problem on Leetcode: https://leetcode.com/problems/monthly-transactions-i/

My question doesn't deal much with the actual problem, all you need to know is that there is a column 'trans_date' that has dates in the YYYY-MM-DD format, and I'd like to convert this date to the YYYY-MM format.

But, if I write:

import pandas as pd

def monthly_transactions(transactions: pd.DataFrame) -> pd.DataFrame:
transactions['trans_date'] = transactions['trans_date'].dt.to_period('M')
return transactions


I get a Runtime error when running on Leetcode:

OverflowError: Maximum recursion level reached
return ujson_dumps(
Line 263 in write (/usr/local/lib/python3.10/dist-packages/pandas/io/json/_json.py)
).write()
Line 210 in to_json (/usr/local/lib/python3.10/dist-packages/pandas/io/json/_json.py)
return json.to_json(
Line 2702 in to_json (/usr/local/lib/python3.10/dist-packages/pandas/core/generic.py)
return func(*args, **kwargs)
Line 333 in wrapper (/usr/local/lib/python3.10/dist-packages/pandas/util/_decorators.py)
json_file = simplejson.loads(result_table.to_json(orient='split'))
Line 48 in main (_driver.py)
main()
Line 62 in <module> (_driver.py)


But, this is fine (it runs and gives the wrong answer):

import pandas as pd

def monthly_transactions(transactions: pd.DataFrame) -> pd.DataFrame:
transactions['trans_date'].dt.to_period('M')
return transactions


I'm pretty confused here, since running this on my own computer and not through Leetcode is fine.

Can anyone else reproduce this behavior and/or suggest how to fix it?

Continue reading...