chenghaibo / dummipy

Categorical variables for pandas DataFrames and scikit-learn

dummipy

Quickstart

$ pip install dummipy

Let it out of the box...

from sklearn.linear_model import LinearRegression
from dummipy import cereal

type(cereal)
# CategoricalDataFrame

cereal.head()
reg = LinearRegression()
reg.fit(cereal[['mfr', 'vitamins', 'fat']], cereal.calories)

Installation

You'll need pandas, but any old version will do the trick. There is no pandas version pegged in the setup.py file so installing dummipy won't mess up your existing sci-py setup.

$ pip install dummipy

Use

Just use it like any old data frame. That's really all there is to it.

import dummipy as dp

df = dp.CategoricalDataFrame({
    "x": range(5),
    "y": ["a", "b", "c", "a", "b"]
})


df = pd.read_csv("foo.csv")
df = dp.CategoricalDataFrame(df)