The aim of this project is to process the Logistic Regression

• docker container run --name classificationapi -p 8081:8000 -d tahapek5454/body_type_classification:latest
• go http://127.0.0.1:8081/docs

• First, you run fitModel
• after that you can predict data

Note: SetData only work your local not in container

Logistic Regression is a classification algorithm commonly used for binary classification problems. This algorithm attempts to separate data points into two different classes by using a linear combination of independent variables.

• Objective: To obtain a binary output (usually 0 and 1).
• Model: A linear function and a sigmoid function that transforms the result into a log-odds ratio.
• Training: The model learns its parameters (weights) by fitting to the training data.

The sigmoid (logistic) function used in Logistic Regression is defined as:

[ \sigma(z) = \frac{1}{1 + e^{-z}} ]

This function transforms any real number (z) into a value between 0 and 1.

The fundamental equation for the Logistic Regression model is:

[ P(Y=1|X) = \frac{1}{1 + e^{-(b + w_1x_1 + w_2x_2 + \ldots + w_nx_n)}} ]

In this equation:

• ( P(Y=1|X) ): Probability of being class 1 for a given input (X).
• ( b ): Bias or intercept term.
• ( w_1, w_2, \ldots, w_n ): Weights.
• ( x_1, x_2, \ldots, x_n ): Input features.

1. Training: Model parameters are trained on the dataset.
2. Prediction: The trained model makes classification predictions for new input features.

• Simple and fast.
• Performs well, especially when there is a linear relationship between features.

• Cannot model non-linear relationships.
• Prone to overfitting, so proper regularization may be needed.
• The dependent variable must be categorical (binary classification).

Logistic Regression is widely used in various fields such as medicine, economics, marketing, and bioinformatics to solve classification problems.