EazyML Explainable AI Template

Define Imports

!pip install --upgrade eazyml-xai
!pip install --upgrade eazyml-automl
!pip install gdown python-dotenv

import os
from eazyml_xai import (
    ez_init,
    ez_explain
)

from eazyml import (
    ez_display_df,
    ez_build_model
)

import gdown
import pandas as pd

from dotenv import load_dotenv
load_dotenv()

True

1. Initialize EazyML

The ez_init function uses the EAZYML_ACCESS_KEY environment variable for authentication. If the variable is not set, it defaults to a trial license.

ez_init(access_key=os.getenv('EAZYML_ACCESS_KEY'))

{'success': True,
 'message': 'Initialized successfully. You may revoke your consent to sharing usage stats anytime. You have exclusive paid access.'}

2. Define Dataset Files and Outcome Variable

gdown.download_folder(id='1EobxYR3pg_Z3Sd4sETfe4aJLAsT98fL2')

# Names of the files that will be used by EazyML APIs
train_file_path = os.path.join('data', 'Heart_Attack_traindata.csv')
test_file_path  = os.path.join('data', 'Heart_Attack_testdata.csv')

# The column name for outcome of interest
outcome = 'class'

3. Dataset Information

The dataset used in this notebook is the Heart Attack Dataset, which is a well-known dataset in machine learning and statistics. It contains data about patients, with several features (such as age, gender, blood pressure levels, and heart-related measurements) to predict the likelihood of a heart attack.

Columns in the Dataset:

• age: The age of the patient, measured in years.
• gender: The gender of the patient, represented as a categorical variable (e.g., 1 = male, 0 = female).
• impulse: Refers to the patient's pulse rate, measured in beats per minute (bpm).
• pressurehight: Refers to systolic blood pressure, the higher number in a blood pressure reading (e.g., 120/80 mmHg).
• pressurelow: Refers to diastolic blood pressure, the lower number in a blood pressure reading (e.g., 120/80 mmHg).
• glucose: A measurement related to the heart, likely referring to potassium (K) concentration.
• kcm: This refer to a measurement related to the heart, related to potassium (K) concentration.
• troponin: A protein found in the heart muscle, measured to assess heart damage (especially after a heart attack).
• class: The target variable, indicating the presence or absence of a condition or disease (e.g., 1 = heart attack, 0 = no heart attack).

3.1 Display the Dataset

Below is a preview of the dataset:

# Load the dataset from the provided file
train = pd.read_csv(train_file_path)

# Display the first few rows of the dataset
ez_display_df(train.head())

	age	gender	impluse	pressurehight	pressurelow	glucose	kcm	troponin	class
0	64	1	66	160	83	160.000000	1.800000	0.012000	negative
1	21	1	94	98	46	296.000000	6.750000	1.060000	positive
2	55	1	64	160	77	270.000000	1.990000	0.003000	negative
3	64	1	70	120	55	270.000000	13.870000	0.122000	positive
4	55	1	64	112	65	300.000000	1.080000	0.003000	negative

4. EazyML Predictive Models

4.1 Reading the Datasets and Dropping Unnecessary Columns

discard_columns = []

# Reading Training Data
train = pd.read_csv(train_file_path)
train = train.drop(columns=discard_columns)

# Reading Test Data
test = pd.read_csv(test_file_path)
test = test.drop(columns=discard_columns)

4.2 Model Training: Several Models Trained

## Build Model
options = {'model_type': 'predictive'}
resp = ez_build_model(train, outcome=outcome, options=options)

4.3 Show Model Performance

ez_display_df(resp['model_performance'])

	Model	Kappa	Accuracy
0	Bagged Decision Trees with Information Gain	0.98	0.99
1	Random Forest with Information Gain	0.98	0.99
2	Gradient Boosting Classifier	0.98	0.99
3	Boosted Decision Trees with InformationGain	0.96	0.98
4	Logistic Regression	0.59	0.80
5	Naive Bayes	0.44	0.70

5. Get Explanations

5.1 Use model_info from ez_build_model

# In extra info, we have model information
model_info = resp["model_info"]

5.2 Get Explanations for 5 Points

options = {'record_number': [1, 6, 7, 8, 9]}
response = ez_explain(train_file_path, outcome, test_file_path, model_info, options=options)

5.3 Display Explanation DataFrame

ex_df = pd.DataFrame([i.values() for i in response['explanations']], columns=response['explanations'][0].keys())
ez_display_df(ex_df)

	record_numbers	prediction	explanation	explainability_score	local_importance
0	1	negative	troponin is 0.01 (that is in range -0.32, 0.01) , kcm is 1.8 (that is less than or equal to 4.95)	97%	{'troponin': 0.65, 'kcm': 0.18, 'pressurehight': 0.09, 'gender': 0.08}
1	6	negative	troponin is 0.0 (that is in range -0.31, 0.01) , kcm is 1.83 (that is less than or equal to 4.95) , gender is 1	96%	{'troponin': 0.67, 'gender': 0.18, 'kcm': 0.16}
2	7	negative	pressurehight is 179.0 (that is more than 157.5) , troponin is 0.0 (that is more than -0.3), gender is not 1	31%	{'pressurehight': 0.38, 'troponin': 0.36, 'kcm': 0.16, 'gender': 0.1}
3	8	positive	kcm is 300.0 (that is more than 256.83) , pressurehight is 214.0 (that is more than 183.01) , troponin is 2.37 (that is less than or equal to 2.37)	25%	{'kcm': 0.7, 'troponin': 0.26}
4	9	negative	troponin is 0.0 (that is in range -0.3, 0.01) , kcm is 2.35 (that is in range -2.14, 6.29)	97%	{'troponin': 0.64, 'kcm': 0.19, 'gender': 0.13}