BrainStrokeClassifier 0.1.3

A Machine Learning Brain Stroke Classfier.

BrainStrokeClassifier API

Functionality of the BrainStrokeClassifier

The package includes a trained Machine Learning model to predict Brain Stroke vulnerability. The model can be accessed as BStClassifier utility. You can use BStClassifier.predict(X) to make predictions. Where X is the input consisting of 7 characteristics:

1. Gender ("Male", "Female", "Other")
2. Age
3. Do you have hypertension (Yes/NO)
4. Do you have any heart disease (Yes/NO)
5. What type of work you do? ("Private job", "Self Employed", "Govt. Job", "Never Worked", "You are a Child")
6. Your average blood glucose level
7. Your Body Mass Index (BMI)

It classifies whether you are vulnerable to or had brain stroke, or not.

Input type

The input can be:

1. NumPy array a. 1D array of shape (7,) b. 2D array of shape (S,7); where S is number of samples
2. Pandas Series of shape (7,)
3. Pandas DataFrame of shape (S,7); where S is number of columns

About the Machine Learning model

The Machine Learning model is a Pipeline of Strandard Scaler and XGBoost Classifier. You can check the notebook where the model is trained through the github

Package Installation

Make sure you have Python installed in your system. Run Following command in the command window.

 !pip install BrainStrokeClassifier

Example Code

# test.py

# Package installation
!pip install BrainStrokeClassifier

# Loading package
from BrainStrokeClassifier import BStClassifier
import numpy as np
import pandas as pd

# Creating some test input

x = np.array([[0,67,0,1,0,228.69,36.6],
             [0,58,1,0,0,87.96,39.2]])

df = pd.DataFrame(x)
prediction = BStClassifier(df)
print(prediction)

For Suggestions

If you have any suggestions or improvements, please fork the package on github