CodeUtility
Making your developer's life easier!
Back to challenges

AI learning path

Learn machine learning by coding it in C#

15 challenges where you write the classic machine learning algorithms yourself, with no libraries: data scaling, linear and logistic regression, the perceptron, k-nearest neighbors, Naive Bayes, decision-tree splits and k-means. Solve them in order in C#; each one runs in your browser with instant tests and an AI review.

Start the first challenge
  1. 1

    Working with data

    Matrix multiplication and feature scaling.

    0/3
    1. 1 Matrix Multiplication Easy Solve in C#
    2. 2 Min-Max Normalization Easy Solve in C#
    3. 3 Z-Score Standardization Easy Solve in C#
  2. 2

    Regression

    Fit a line with a formula and with gradient descent, then predict probabilities.

    0/4
    1. 1 Simple Linear Regression (Least Squares) Medium Solve in C#
    2. 2 Linear Regression with Gradient Descent Medium Solve in C#
    3. 3 Logistic Regression Prediction Easy Solve in C#
    4. 4 Perceptron Learning Algorithm Medium Solve in C#
  3. 3

    Classification

    Softmax, cross-entropy, nearest neighbors, Naive Bayes and decision-tree splits.

    0/6
    1. 1 Softmax Function Easy Solve in C#
    2. 2 Categorical Cross-Entropy Loss Easy Solve in C#
    3. 3 K-Nearest Neighbors Classifier Medium Solve in C#
    4. 4 Naive Bayes Classifier Medium Solve in C#
    5. 5 Gini Impurity Easy Solve in C#
    6. 6 Information Gain Medium Solve in C#
  4. 4

    Similarity and clustering

    Cosine similarity and k-means clustering.

    0/2
    1. 1 Cosine Similarity Easy Solve in C#
    2. 2 K-Means Clustering Medium Solve in C#