pandas/apply-known-function
| Rule | apply-known-function |
|---|---|
| Pack | pandas |
| Severity | warning |
| Fix | unsafe, with xpd check --fix --unsafe-fixes |
| Library versions | every version |
Passing a function such as len, pd.isna or pd.to_datetime to apply calls it once for every value. Call the column method or pass the whole column to the function.
lengths = df["name"].apply(len)lengths = df["name"].str.len()What the rule reports
Section titled “What the rule reports”The rule reports df["s"].apply(len), df["b"].apply(pd.isna), df["raw"].apply(pd.to_datetime) and df["b"].apply(lambda v: round(v, 2)). It reports only a column taken by a text label, because on a frame apply works per column.
Each fix is unsafe, and the table gives the reason.
| Fix | What can differ |
|---|---|
pd.to_datetime(df["raw"]) |
a call per value parses each text on its own, and the whole column must share one format |
df["b"].isna() |
on a column of lists, pd.isna per value returned an array per cell |
df["b"].round(2) |
Python and NumPy round some halfway values differently |
df["s"].str.len() |
it raises on a column that is not text |
How to fix it
Section titled “How to fix it”xpd check --fix --unsafe-fixes applies the fixes of this rule. Check the result, because an unsafe fix can change what the code does.
The unsafe fix writes df["s"].str.len(), which raises on a column that is not text, df["b"].isna(), pd.to_datetime(df["raw"]), df["b"].round(2).
Message
Section titled “Message”xpd prints this message with the issue. A name that starts with $ stands for code from your file.
pd.to_datetime takes the whole column, and calling it per value is much slower. Write pd.to_datetime($S).Browse the other rules of the pandas pack.