In feature selection, tanH refers to Hyperbolic Tangent Activation. This function is used Neural network Architecture to transform feature values into a range between -1 and 1. It means it centers the data around 0. It improves the feature selection in deep learning for feature ranking or dimensionality reduction
Load libraries:
Read data
The depenedent variable ‘outcome’ is dichotomous. either 1 or 0.
Drop unwanted column:
Separate features(X) and target(y)
Sequential Model function
Training:
.
.
An epoch refers to a complete pass through the entire training dataset during one cycle of model training. In this process, every sample in the dataset is fed through the model, and its weights and biases are updated based on the calculated loss or error. In deep learning, the dataset is often divided into smaller subsets. Aim is to minimze the loss.
Performances:
