Dataset Information
The data set consists of a total of 11 features. The target variable is a binary value indicating whether or not a cardiovascular disease is present in the subject.
Feature Information
Age | Objective Feature | age | int (days) |
Height | Objective Feature | height | int (cm) |
Weight | Objective Feature | weight | float (kg) |
Gender | Objective Feature | gender | categorical code | 1: Women, 2: Men |
Systolic blood pressure | Examination Feature | ap_hi | int |
Diastolic blood pressure | Examination Feature | ap_lo | int |
Cholesterol | Examination Feature | cholesterol | 1: normal, 2: above normal, 3: well above normal |
Glucose | Examination Feature | gluc | 1: normal, 2: above normal, 3: well above normal |
Smoking | Subjective Feature | smoke | binary |
Alcohol intake | Subjective Feature | alco | binary |
Physical activity | Subjective Feature | active | binary |
Presence or absence of cardiovascular disease | Target Variable | cardio | binary |
The model is densely connected feedforward Neural Network. The Jupyter Notebook has a more extensive explanation of the following info.
relu
The vanishing gradient problem: the gradients of some activation functions becoming increasingly smaller as the number of hidden layers increases. This is problematic because the parameters in the neural network won't be tuned effectively.
ReLU overcomes this problem.
sigmoid
Activation function used in the output layer. Effective for classification problems.
Binary Crossentropy
Cross entropy is calculated by finding the predicted probability (the probability of classification being correct based on given data). Larger predicted probabilities yeild lower losses. Binary Cross Entropy loss is a sigmoid activiation function along with the cross entropy loss function.
Adam Optmizier
Calculates adaptive learning rates for each parameter.