In this section we are planning to predict diabetes by specific parameters and the referred parameters are described:
This dataset is originally from the National Institute of Diabetes and Digestive and Kidney Diseases. The objective of the dataset is to diagnostically predict whether a patient has diabetes, based on certain diagnostic measurements included in the dataset. Several constraints were placed on the selection of these instances from a larger database. In particular, all patients here are females at least 21 years old of Pima Indian heritage.2 From the data set in the (.csv) File We can find several variables, some of them are independent (several medical predictor variables) and only one target dependent variable (Outcome).
Pregnancies: To express the Number of pregnancies
Glucose: To express the Glucose level in blood
BloodPressure: To express the Blood pressure measurement
SkinThickness: To express the thickness of the skin
Insulin: To express the Insulin level in blood
BMI: To express the Body mass index
DiabetesPedigreeFunction: To express the Diabetes percentage
Age: To express the age
Outcome: To express the final result 1 is Yes and 0 is No
Required libraries and some settings for this section are:
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
pd.set_option("display.max_columns", None)
pd.set_option("display.max_rows", None)
pd.set_option("display.width", 500)
pd.set_option("display.float_format", lambda x: "%.4f" % x)
from sklearn.model_selection import train_test_split, cross_validate, GridSearchCV, cross_val_score
from sklearn.ensemble import RandomForestClassifier
from sklearn.metrics import accuracy_score, confusion_matrix, classification_report
import joblib
import warnings
warnings.filterwarnings("ignore")
First, we import the dataset diabetes.csv into the pandas DataFrame.
As we want to check the data to have a general opinion about it, we create and use a function called check_df(dataframe, head=5, tail=5) that prints the referred functions:
dataframe.head(head)
dataframe.tail(tail)
dataframe.shape
dataframe.dtypes
dataframe.size
dataframe.isnull().sum()
dataframe.describe([0, 0.01, 0.05, 0.10, 0.20, 0.30, 0.40, 0.50, 0.60, 0.70, 0.80, 0.90, 0.95, 0.99, 1]).T
After checking the data frame, we need to define and separate columns as categorical and numerical. We define a function called grab_col_names for separation that benefits from multiple list comprehensions as follows:
cat_cols = [col for col in dataframe.columns if str(dataframe[col].dtypes) in ['category', 'object', 'bool']]
num_but_cat = [col for col in dataframe.columns if dataframe[col].nunique() < cat_th and dataframe[col].dtypes in ['uint8', 'int64', 'int32', 'float64']]
cat_but_car = [col for col in df.columns if df[col].nunique() > car_th and str(df[col].dtypes) in ['object', 'category']]
cat_cols = cat_cols + num_but_cat
num_cols = [col for col in dataframe.columns if dataframe[col].dtypes in ['uint8', 'int64', 'float64']]
num_cols = [col for col in num_cols if col not in cat_cols]
cat_th and car_th are the threshold parameters to decide the column type.
Categorical Columns:
- Outcome
Numerical Columns:
- Pregnancies
- Glucose
- Blood Pressure
- Skin Thickness
- Insulin
- BMI
- Diabetes Pedigree Function
- Age
To summarize and visualize the referred column we create two other functions called cat_summary and num_summary.
For example, categorical column Outcome:
############### Outcome ###############
| Outcome | Outcome Nr | Ratio |
|---|---|---|
| 0 | 500 | 65.1042 |
| 1 | 268 | 34.8958 |
Another example is, the numerical column Pregnancies:
############### Pregnancies ###############
| Process | Result |
|---|---|
| count | 768.0000 |
| mean | 3.8451 |
| std | 3.3696 |
| min | 0.0000 |
| 1% | 0.0000 |
| 5% | 0.0000 |
| 10% | 0.0000 |
| 20% | 1.0000 |
| 30% | 1.0000 |
| 40% | 2.0000 |
| 50% | 3.0000 |
| 60% | 4.0000 |
| 70% | 5.0000 |
| 80% | 7.0000 |
| 90% | 9.0000 |
| 95% | 10.0000 |
| 99% | 13.0000 |
| max | 17.0000 |
Name: Pregnancies, dtype: float64
With the help of a for loop we apply these functions to all columns in the data frame.
We create another plot function called plot_num_summary(dataframe) to see the whole summary of numerical columns due to the high quantity of them:
We create another function called target_summary_with_cat(dataframe, target, categorical_col) to examine the target by categorical features.
For instance Glucose Feature
################ Outcome --> Glucose #################
| Outcome | Glucose |
|---|---|
| 0 | 109.9800 |
| 1 | 141.2575 |
To analyze correlations between numerical columns we create a function called correlated_cols(dataframe):
We check the data to designate the missing values in it, dataframe.isnull().sum():
| Feature | Missing Value |
|---|---|
| Pregnancies | 0 |
| Glucose | 0 |
| BloodPressure | 0 |
| SkinThickness | 0 |
| Insulin | 0 |
| BMI | 0 |
| DiabetesPedigreeFunction | 0 |
| Age | 0 |
| Outcome | 0 |
dtype: int64
Even though it seems there are no missing values, some of the zero '0' values mean that they are actually missing;
To clearly observe the missing values, we are creating a list called zero_columns via lis comprehension.
Actual result of the missing values:
| Feature | Missing Value |
|---|---|
| Pregnancies | 0 |
| Glucose | 5 |
| BloodPressure | 35 |
| SkinThickness | 227 |
| Insulin | 374 |
| BMI | 11 |
| DiabetesPedigreeFunction | 0 |
| Age | 0 |
| Outcome | 0 |
Missing Values Table:
We fill the missing values with the median value of the referring feature.
At last, we create our model and see the results:
******************** RF Model Results ********************
Accuracy Train: 1.000
Accuracy Test: 0.749
R2 Train: 1.000
R2 Test: 0.749
Cross Validation Train: 0.756
Cross Validation Test: 0.719
Cross Validation (Accuracy): 0.764
Cross Validation (F1): 0.635
Cross Validation (Precision): 0.692
Cross Validation (Recall): 0.590
Cross Validation (ROC Auc): 0.831
Via joblib we can save and/or load our model:
def load_model(pklfile):
model_disc = joblib.load(pklfile)
return model_disc
model_disc = load_model("rf_model.pkl")
Now we can make predictions with our model:
X = df.drop("Outcome", axis=1)
x = X.sample(1).values.tolist()
model_disc.predict(pd.DataFrame(X))[0]
1
sample2 = [3, 165, 68, 38, 115, 30, 0.58, 32]
model_disc.predict(pd.DataFrame(sample2).T)[0]
1
To have better predictions we tune our model and the results are:
Fitting 5 folds for each of 24 candidates, totalling 120 fits
******************** RF Model Results ********************
Accuracy Train: 0.931
Accuracy Test: 0.913
R2 Train: 0.931
R2 Test: 0.913
Cross Validation Train: 0.769
Cross Validation Test: 0.741
Cross Validation (Accuracy): 0.760
Cross Validation (F1): 0.643
Cross Validation (Precision): 0.677
Cross Validation (Recall): 0.620
Cross Validation (ROC Auc): 0.834
def load_model(pklfile):
model_disc = joblib.load(pklfile)
return model_disc
model_disc = load_model("rf_model_tuned.pkl")
X = df.drop("Outcome", axis=1)
x = X.sample(1).values.tolist()
model_disc.predict(pd.DataFrame(X))[0]
1
sample2 = [3, 165, 68, 38, 115, 30, 0.58, 32]
model_disc.predict(pd.DataFrame(sample2).T)[0]
1