zedda.fix()

Generate copy-paste pandas fix code, or apply fixes in place.

zedda.fix(
    path,
    apply=False,
    sample_size=None,
    correlate=False,
) -> None | pandas.DataFrame

Generate copy-paste pandas fix code grouped by issue type, or apply the fixes in place and return the cleaned DataFrame.

Arguments

Argument Type Default Description
path str / Path / DataFrame (required) Input dataset
apply bool False If True, execute fixes on a DataFrame and return it. If False, print code only.
sample_size int or None None Sample size in rows
correlate bool False Force O(N²) correlation

Returns

  • apply=False (default) → None. Prints the fix code to the terminal.
  • apply=True → pandas.DataFrame. The cleaned DataFrame (the original file on disk is not modified).

Generated code structure

The fix code is grouped by issue type:

Issue category Generated code
Missing Values df["col"].fillna(df["col"].median(), inplace=True) for numeric, fillna(df["col"].mode()[0]) for strings
Outliers df["col"] = df["col"].clip(upper=df["col"].quantile(0.99))
ID Columns df.drop(columns=["col"], inplace=True)
High-Cardinality Strings df["col"] = pd.factorize(df["col"])[0]
Constant Columns df.drop(columns=["col"], inplace=True)

Example

import zedda as zd

# Print fix code only
zd.fix("data.csv")

# Apply and get the cleaned DataFrame
cleaned = zd.fix("data.csv", apply=True)
print(cleaned.shape)

Difference from clean()

Aspect fix(apply=True) clean()
Writes a backup file No Yes ({path}.zedda-backup)
Writes a JSON audit trail No Yes
Returns a DataFrame Yes Yes (if input is DataFrame or output=None)
Writes a cleaned file to disk No Yes
Has an undo() helper No Yes (clean.undo())

Use fix() when you want to review and edit the code. Use clean() when you want safe, audited auto-cleaning.

See also