Building Neural Networks from Scratch
4.0
(7)
47 learners
What you'll learn
This course includes
- 17.5 hours of video
- Certificate of completion
- Access on mobile and TV
Course content
1 modules • 35 lessons • 17.5 hours of video
Building Neural Networks from Scratch
35 lessons
• 17.5 hours
Building Neural Networks from Scratch
35 lessons
• 17.5 hours
- Lecture 1 - Neural Network from Scratch: Coding Neurons and Layers 28:37
- Lecture 2 - The beauty of numpy and dot product in coding neurons and layers 40:21
- Lecture 3 - Coding multiple neural network layers and stacking them together 13:08
- Lecture 4 - Implementing the Dense Layer Class in Python 25:24
- Lecture 5 - Broadcasting and array summation in Python 25:28
- Lecture 6 - Coding Neural Network Activation Functions from scratch 43:16
- Lecture 7 - Coding one neural network forward pass in Python (no loss) 14:33
- Lecture 8 - Coding the cross entropy loss in Python (from scratch) 44:03
- Lecture 9 - Introduction to Optimization in Neural Network training 27:06
- Lecture 10 - Partial Derivatives and Gradient in Neural Networks 24:56
- Lecture 11 - Understand Chain Rule: The backbone of Neural Networks 21:08
- Lecture 12 - Backpropagation from scratch on a single neuron 34:01
- Lecture 13 - Backpropagation through an entire layer of neurons: from scratch 30:11
- Lecture 14 - Role of matrices in backpropagation 39:28
- Lecture 15 - Finding derivatives of inputs in backpropagation and why we need them 28:28
- Lecture 16 - Coding Backpropagation building blocks in Python 22:59
- Lecture 17 - Backpropagation on the ReLU activation class 12:45
- Lecrure 18 - Implementing backpropagation on the cross entropy loss function 26:31
- Lecture 19 - Combined backpropagation on softmax activation and cross entropy loss 28:08
- Lecture 20 - Build the entire backpropagation pipeline for neural networks | No PyTorch, Tensorflow 22:10
- Lecture 21 - Coding the entire neural network forward backward pass in Python 19:34
- Lecture 22 - Coding Optimizers in Neural Networks: Gradient Descent 31:03
- Lecture 23 - Learning Rate Decay in Neural Network Optimization 20:28
- Lecture 24 - Momentum in training neural networks 26:54
- Lecture 25 -Coding the ADAGRAD optimizer for Neural Network training 38:44
- Lecture 26 - Coding the RMSProp Optimizer with Neural Network training 32:00
- Lecture 27 -Coding the ADAM optimizer for neural networks 45:20
- Neural Networks in 100 minutes: Coded from scratch 01:41:53
- Lecture 28 -Neural network testing, generalization and overfitting 18:18
- Lecture 29 -K-fold cross validation in Neural Networks 14:35
- Lecture 30 - L1/L2 Regularization to avoid neural network overfitting 42:20
- Lecture 31 - Dropout layers in neural networks - Full code 32:15
- Hands on Neural Networks Project: MNIST Fashion Dataset 46:35
- Hands on Neural Networks Project: California Housing Dataset 32:46
- 20 minutes summary: Building, training and testing neural networks from scratch 20:35
Take this course with you
Download the app to continue lessons from your phone anytime.
