PIMA diabetes looking into different sampling options

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns

from jmspack.utils import (apply_scaling,
                          JmsColors)
from imblearn.over_sampling import (SMOTE,
                                    ADASYN,
                                    BorderlineSMOTE,
                                    SVMSMOTE,
                                    KMeansSMOTE
                                   )

if "jms_style_sheet" in plt.style.available:
    plt.style.use("jms_style_sheet")

df = pd.read_csv("diabetes.csv")

df.head()

	Pregnancies	Glucose	BloodPressure	SkinThickness	Insulin	BMI	DiabetesPedigreeFunction	Age	Outcome
0	6	148	72	35	0	33.6	0.627	50	1
1	1	85	66	29	0	26.6	0.351	31	0
2	8	183	64	0	0	23.3	0.672	32	1
3	1	89	66	23	94	28.1	0.167	21	0
4	0	137	40	35	168	43.1	2.288	33	1

target = "Outcome"

df.describe()

	Pregnancies	Glucose	BloodPressure	SkinThickness	Insulin	BMI	DiabetesPedigreeFunction	Age	Outcome
count	768.000000	768.000000	768.000000	768.000000	768.000000	768.000000	768.000000	768.000000	768.000000
mean	3.845052	120.894531	69.105469	20.536458	79.799479	31.992578	0.471876	33.240885	0.348958
std	3.369578	31.972618	19.355807	15.952218	115.244002	7.884160	0.331329	11.760232	0.476951
min	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.078000	21.000000	0.000000
25%	1.000000	99.000000	62.000000	0.000000	0.000000	27.300000	0.243750	24.000000	0.000000
50%	3.000000	117.000000	72.000000	23.000000	30.500000	32.000000	0.372500	29.000000	0.000000
75%	6.000000	140.250000	80.000000	32.000000	127.250000	36.600000	0.626250	41.000000	1.000000
max	17.000000	199.000000	122.000000	99.000000	846.000000	67.100000	2.420000	81.000000	1.000000

Mask 0 default values (seen in EDA)

df = (df.drop([target, "Pregnancies"], axis=1)
       .replace(0, np.nan)
       .merge(df[[target, "Pregnancies"]], left_index=True, right_index=True)
      .dropna()
     )

Plot the amount of rows in each side of the target

Looks like the target is imbalanced so this needs to be taken into account

_ = sns.countplot(x=df[target])
print("Amount in each outcome")
df[target].value_counts()

Amount in each outcome

0    262
1    130
Name: Outcome, dtype: int64

There are two main ways to take this imbalance into account, either by (re)-sampling the data to make the outcome amount equal, or by using a classifier which takes the imbalance into account in the model (usually known as a bagging classifier)

feature_list = ["Glucose", "BMI"]

sampling_options_dict = {"SMOTE": SMOTE(sampling_strategy="not majority",
                                        random_state=42,
                                        n_jobs=2),
                         "BorderlineSMOTE": BorderlineSMOTE(sampling_strategy="not majority",
                                        random_state=42,
                                        n_jobs=2),
                         "SVMSMOTE": SVMSMOTE(sampling_strategy="not majority",
                                        random_state=42,
                                        n_jobs=2),
                         "KMeansSMOTE": KMeansSMOTE(sampling_strategy="not majority",
                                        random_state=42,
                                        n_jobs=2),
                         "ADASYN": ADASYN(sampling_strategy="not majority",
                                        random_state=42,
                                        n_jobs=2),
                        }

X = df.drop(target, axis=1)
y = df[target]

for samp_opt in sampling_options_dict:
    print(samp_opt)

    sm = sampling_options_dict[samp_opt]
    X_res, y_res = sm.fit_resample(X, y)
    print(X.shape, y.shape, X_res.shape, y_res.shape)

    df_res = pd.concat([X_res, y_res], axis=1)

    _ = sns.scatterplot(data=df_res,
                       x=feature_list[0],
                       y=feature_list[1],
                       color=JmsColors.DARKBLUE,
                       style=target)
    _ = sns.scatterplot(data=df,
                       x=feature_list[0],
                       y=feature_list[1],
                       color=JmsColors.GREENYELLOW,
                       style=target,
                       legend=False)
    _ = plt.plot(df_res[feature_list[0]].min(),
                    df_res[feature_list[1]].min(),
                    c=JmsColors.DARKBLUE,
                    label = "resampled data")
    _ = plt.plot(df_res[feature_list[0]].min(),
                    df_res[feature_list[1]].min(),
                    c=JmsColors.GREENYELLOW,
                    label = "original data")
    _ = plt.legend()
    _ = plt.title(f"Data following resampling: {samp_opt}")
    _ = plt.show()

SMOTE
(392, 8) (392,) (524, 8) (524,)

BorderlineSMOTE
(392, 8) (392,) (524, 8) (524,)

SVMSMOTE
(392, 8) (392,) (524, 8) (524,)

KMeansSMOTE
(392, 8) (392,) (525, 8) (525,)

ADASYN
(392, 8) (392,) (525, 8) (525,)