AI From Scratch

The purpose of this repository is to understand the mathematics and implementation details behind Machine Learning algorithms before relying on high-level frameworks.

---

Current Implementations

Algorithm	Status
Linear Regression	✅
Logistic Regression	✅
K-Nearest Neighbors (KNN)	✅
Perceptron	✅

---

Planned Implementations

Algorithm	Status
k-means clustering	⏳
Decision Tree	⏳
Random Forest	⏳
Naive Bayes	⏳
Support Vector Machine (SVM)	⏳
Neural Networks	⏳
Convolutional Neural Networks (CNN)	⏳
Recurrent Neural Networks (RNN)	⏳

---

-i've written an article about Logistic Regression and the mathematics behind it. if u like u can take a look at it!

-https://vrgl.ir/TIDBe

---

Implemented Algorithms

Linear Regression

Linear Regression predicts continuous numerical values by fitting a straight line to the training data.

Mathematical Model

$$ \hat{y}=wx+b $$

Loss Function

Huber Loss

Optimizer

Gradient Descent

---

Example Result

---

Logistic Regression

Logistic Regression is a binary classification algorithm that predicts probabilities using the Sigmoid activation function.

Linear Model

$$ z=wx+b $$

Activation Function

Sigmoid

$$ \sigma(z)=\frac1{1+e^{-z}} $$

Loss Function

Binary Cross Entropy (BCE)

Optimizer

Gradient Descent

---

Example Result

---

Other Models pic (knn, perceptron)