8086ML

This repository contains implementations of basic machine learning algorithms written entirely in Assembly language. These implementations aim to demonstrate fundamental concepts of machine learning in a highly constrained programming environment.

Decision Tree

Method Used: The Decision Tree code predicts whether it will rain based on the temperature and humidity inputs. It uses simple rules (greater than or less than a certain value) to decide the outcome.

Code Explanation:

• Input: Temperature and Humidity
• Process: Checks if the temperature is above or below 70 degrees. Based on that, it further checks humidity levels to predict rain.
• Output: Predicts "It will rain" or "It will not rain."

k-Nearest Neighbors (k-NN)

Method Used: k-NN is used here to predict prices based on area. The algorithm looks at 'k' closest data points (based on distance) to predict the price.

Code Explanation:

• Input: Area of a property.
• Process: Calculates distances between input area and known data points, selects the nearest ones, and uses their average prices for prediction.
• Output: Predicts a price based on the area.Due to memory constrains it divides the price by 1000.

Linear Regression

Method Used: Linear Regression is implemented to estimate prices based on area. It finds a line (mathematically) that best fits the data points.

Code Explanation:

• Input: Area of a property.
• Process: Computes a line that best describes the relationship between area and price by minimizing the distance between the line and the data points.
• Output: Gives an estimated price for a given area.

Logistic Regression

Method Used: Logistic Regression is used to classify images as belonging to one of two classes. This model uses a simplified step function as an activation instead of the typical sigmoid function.

Code Explanation:

• Input: Pixel values of a 4x4 image.
• Process: Each pixel is multiplied by a corresponding weight and summed up, including a bias. A step function checks if the result is above or below zero to classify the image.
• Output: Classifies the image into class 0 or class 1.

---

"...are you mad? ~William Shakespeare"

Feel free to explore the code, suggest improvements, or fork the repository!!