Velamuvi AIAcademy
Deep Learning

Artificial Neural Networks

Build a neural network from scratch in NumPy, then again in PyTorch, and understand every number in it.

  • Intermediate
  • About 12 hours
  • 4 modules
  • 14 lessons
  • Free

This course is not released yet. We publish courses one at a time, each one complete. The full syllabus is below.

What you will learn

  • ArchitectureNetwork Architecture, Weights and Biases, Layers: Input, Hidden, Output and more
  • TrainingThe Training Loop, Batches and Epochs, Initialising Weights and more
  • RegularisationDropout, Weight Decay, Batch Normalisation and more
  • ProjectProject: Neural Network From Scratch, Project: Handwritten Digit Recognition

Course syllabus

Not released yet · full syllabus below

  1. 1

    Architecture

    4 lessons
    1. 1Network ArchitectureSoon
    2. 2Weights and BiasesSoon
    3. 3Layers: Input, Hidden, OutputSoon
    4. 4Counting ParametersSoon
  2. 2

    Training

    4 lessons
    1. 5The Training LoopSoon
    2. 6Batches and EpochsSoon
    3. 7Initialising WeightsSoon
    4. 8Vanishing and Exploding GradientsSoon
  3. 3

    Regularisation

    4 lessons
    1. 9DropoutSoon
    2. 10Weight DecaySoon
    3. 11Batch NormalisationSoon
    4. 12Early StoppingSoon
  4. 4

    Project

    2 lessons
    1. 13Project: Neural Network From ScratchSoon
    2. 14Project: Handwritten Digit RecognitionSoon

Recommended next