import pandas as pd
from sklearn.compose import make_column_transformer
from sklearn.preprocessing import OneHotEncoder, StandardScaler
from sklearn.dummy import DummyClassifier
from sklearn.linear_model import LogisticRegression
from sklearn.model_selection import cross_validate
from sklearn.pipeline import make_pipeline
[docs]def classifier(train_df, target_col, numeric_feats = None, categorical_feats = None, cv = 5):
"""This function preprocess the data, fit baseline model(dummyclassifier) and logistic regression with default setups to provide data scientists
easy access to the common models results(scores).
Parameters
----------
train_df : pandas.DataFrame
The clean train data which includes target column.
target_col : str
The column of the train data that has the target values.
numeric_feats = list
The numeric features that needs to be considered in the model. If the user enters an empty list, the function will use all numeric columns.
categorical_feats : list
The categorical features that needs to be considered in the model.
cv : int, optional
The number of folds on the data for train and validation set.
Returns
-------
Data frame
A data frame that includes test scores and train scores for each model.
Examples
-------
>>> classifier(train_df, target_col = 'target', numerical_feats = [], categorical_features = [])
>>> classifier(train_df, target_col = 'target', numeric_feats = ['danceability', 'loudness'], categorical_feats=['genre'], cv=10)
"""
if (not(isinstance(train_df , pd.core.frame.DataFrame))):
raise TypeError("Invalid function input. Please enter a data frame")
if (not (train_df.isna().sum().sum() == 0)):
raise ValueError("Invalid function input. Please pass a clean pandas data frame")
if not(isinstance(numeric_feats , list)):
raise TypeError("Numeric Features should be passed as a list")
if not(isinstance(categorical_feats , list)):
raise TypeError("Categorical Features should be passed as a list")
if not(isinstance(target_col , str)):
raise TypeError("Target column must be passed as a string")
X_train = train_df.drop(columns=target_col, axis=1)
y_train = train_df[target_col]
if not isinstance(numeric_feats, list):
raise TypeError("The numeric features have to be entered as a list")
if not isinstance(categorical_feats , list):
raise TypeError("The categorical features have to be entered as a list")
if numeric_feats == None or numeric_feats==[]:
numeric_feats = train_df.select_dtypes(include='number').columns.tolist()
if categorical_feats == None or categorical_feats ==[]:
categorical_feats = train_df.select_dtypes(exclude='number').columns.tolist()
preprocessor = make_column_transformer(
(StandardScaler(), numeric_feats),
(OneHotEncoder(), categorical_feats))
dummy = DummyClassifier()
lr = make_pipeline(preprocessor, LogisticRegression())
results = pd.Series(dtype='float64')
models = {"DummyClassifier": dummy, "LogisticRegression" : lr}
for model in models :
scores = cross_validate(models[model], X_train, y_train, return_train_score = True,cv = cv)
mean_scores = pd.DataFrame(scores).mean().to_frame(model)
results = pd.concat([results, mean_scores], axis = 1)
results = results.drop(columns = 0, axis=1)
return results