sklego/dict-mapper-not-int
| Rule | dict-mapper-not-int |
|---|---|
| Pack | scikit-lego |
| Severity | error |
| Fix | none, edit by hand |
| Library versions | every version |
DictMapper casts every mapped value to int, so a float loses its fraction without a warning. Map to whole numbers, or map the column with pandas Series.map before the pipeline.
mapper = DictMapper({"low": 0.5, "high": 1.7}, default=-1)mapper = DictMapper({"low": 0, "high": 1}, default=-1)weights = frame["level"].map({"low": 0.5, "high": 1.7}).fillna(-1)What the rule reports
Section titled “What the rule reports”The rule reports DictMapper with a dict literal that maps to a float with a fraction or to text,
and a default that is such a float, text or NaN. DictMapper casts every mapped value and the
default to int.
How to fix it
Section titled “How to fix it”xpd has no automatic fix, so make the change by hand.
| Mapping | Result | Reported |
|---|---|---|
{"a": 0.5, "b": 1.7} with default=-1 |
0, 1 and -1, with no warning | yes |
| a text value | ValueError "invalid literal for int() with base 10" | yes |
default=np.nan |
ValueError "cannot convert float NaN to integer" | yes |
a float with no fraction, such as 1.0 |
the same value | no |
{"a": "1"} with default=0, a string that holds a whole number |
1 | no |
Message
Section titled “Message”xpd prints this message with the issue. A name that starts with $ stands for code from your file.
DictMapper casts every mapped value and the default to int, so a float loses its fraction without a warning. {"a": 0.5, "b": 1.7} came back as 0 and 1. Map to whole numbers, or map the column with pandas Series.map before the pipeline.Browse the other rules of the scikit-lego pack.