[FIXED] How do I know what have been changed by pd.to_datetime?

Issue

Assume it is a large database load from a csv, I don't know there is a "Hello" in date column.

df = {"No.": [1, 2, 3, 4],
      "Date": ["17/4/2022", "18/4/2022", "Hello", "19/4/2022"]}

Output:

No. Date
1   17/4/2022
2   18/4/2022
3   Hello
4   19/4/2022

After df["Date"] = pd.to_datetime(df["Date"], infer_datetime_format=True, errors="coerce")

The "Hello" become "NaT".

Output:

No. Date
1   17/4/2022
2   18/4/2022
3   NaT
4   19/4/2022

How do I know there was a "Hello"?

How do I know what value have been changed by pd.to_datetime?

Solution

As you mentioned you want to know the changes and get only strings. You can do it this way:

import pandas as pd

df = pd.DataFrame({"No.": [1, 2, 3, 4],
      "Date": ["17/4/2022", "18/4/2022", "Hello", "19/4/2022"]})
df['Changes'] = df['Date'].apply(lambda x: x if x.isalpha() else "No Change")

Output:

   No.       Date    Changes
0    1  17/4/2022  No Change
1    2  18/4/2022  No Change
2    3      Hello      Hello
3    4  19/4/2022  No Change

Answered By - Abhyuday Vaish