如何取消Pandas DataFrame列的科学计数法格式

Got it, let's figure out why your current approach didn't work and get that accountnumber column looking right. The pd.options.display.float_format you tried only changes how data shows up in your notebook/console—it doesn't fix the underlying data type or values, which is why you didn't see the effect you wanted.

Here are a few reliable solutions tailored to different scenarios:

Solution 1: Read the column as a string upfront (most foolproof) #

When loading your CSV, tell Pandas to treat accountnumber as a string instead of letting it auto-convert to a float. This preserves the original numeric format entirely, so scientific notation never becomes an issue:

df = pd.read_csv('data.csv', dtype={'accountnumber': str})

This is the best approach if you don't need to do math on the account numbers (which you probably don't!).

Solution 2: Convert an existing float column to integer strings #

If you've already loaded the data and need to fix the column retroactively:
Your values look like integers (e.g., 4.11889000e+11 translates to 411889000000 as a whole number), so first convert the float to an integer to drop the decimal part, then turn it into a string to lock in the format:

# Convert float to integer (this handles the scientific notation conversion)
df['accountnumber'] = df['accountnumber'].astype(int)
# Convert to string to avoid any accidental re-formatting later
df['accountnumber'] = df['accountnumber'].astype(str)

If you're worried about edge cases (like non-integer values slipping in), use safer conversion with pd.to_numeric:

df['accountnumber'] = pd.to_numeric(df['accountnumber'], downcast='integer').astype(str)

Solution 3: Fix display without changing the underlying data #

If you only care about how the column looks when you print it, not the data type itself, adjust your float format to show whole numbers instead of 3 decimal places:

# Apply to all float columns globally
pd.options.display.float_format = '{:.0f}'.format

# Or target just the accountnumber column
df['accountnumber'] = df['accountnumber'].apply(lambda x: '{:.0f}'.format(x))

This will render the values as plain whole numbers, though the data might still be stored as a float in the background.

内容的提问来源于stack exchange，提问作者iprof0214