Introduction to Deep Belief Network (DBN) with Python 2023

Name: Introduction to Deep Belief Network (DBN) with Python 2023
Rating: 4.3 (24 reviews)

Deep Belief Network, Bayesian Belief Network, Restricted Boltzmann Machines, Training DBNs.

Created byHoang Quy La

Last updated 1/2023

English

English [Auto],

What you'll learn

Deep Belief Network (DBN)
Restricted Boltzmann Machines (RBMs)
Contrastive Divergence (CD-k) algorithm
Training DBNs
Fine-tuning
Bayesian Belief Networks (BBNs)

Course content

5 sections • 17 lectures • 2h 13m total length

Course Structure1:55
Explore the course structure and its modules, including the introduction to the deep belief network, a lighter recommender system, supervised learning with bilinear work, and unsupervised learning with the DBN.
IMPORTANT NOTES PLEASE DO NOT SKIP1:00
Discover important notes in Google Colab, where every key concept appears in the Colab notebook. Download the notebook to read the transcript and understand the course without watching the videos.
Overview of DBNs4:43
Provide an overview of deep belief networks, describe their multi-layer stochastic latent variables that capture data correlations, and note their use in image recognition, speech recognition, and video sequences.
Introduction to BBNs Part 18:37
Explore how Bayesian belief networks describe conditional dependencies between random variables and enable future predictions, illustrated with a two-variable example of smiling and sunny weather.
Introduction to BBNs Part 24:16
Examine two information propagation modes in Bayesian belief networks. Follow predictive propagation as information flows along arrows from causes to effects, and trace evidence back through retrospective evaluation.
Introduction to RBNs7:10
Explore restricted Boltzmann machines (rbms) and their unsupervised learning for collaborative filtering, feature extraction, topic modeling, and dimensionality reduction; describe visible and hidden layers and input reconstruction.
Steps to train RBNs5:59
Learn the two-step process to train an RBM, including sampling and contrastive divergence, to predict hidden values, reconstruct inputs, and reveal how input data relates to key features.

Introduction to RBM recommender system, importing libraries and loading dataset10:32
Import pandas, numpy, tensorflow, and matplotlib, and build an rbm recommender system by loading and aligning the movie and rating data in Colab.
Normalizing the data9:21
Group users by id and normalize ratings to the 0–1 interval, then prepare training data and configure visible and hidden units for a deep belief network in TensorFlow.
Gibb's sampling Implementation13:41
Explore Gibbs sampling for deep belief networks by implementing hidden and visible activations, reconstructing inputs from hidden activations, and updating weights and biases with a learning rate and gradients.
RBM recommender system final implementation and showing the result14:42
Implement a final RBM based recommender system in Python, initialize weights and biases, train over multiple epochs with a TensorFlow session, and visualize training error to show results.

Unsupervised Learning with Deep belief Network Implementation part 117:39
Implement unsupervised learning with a deep belief network in python. Clone a dbn, load a 400-sample mnist subset, and train through three hidden layers with constructive divergence.
Unsupervised Learning with Deep belief Network Part 25:42
Explore unsupervised learning with a deep belief network by completing pre-training and fine-tuning, sample from the last hidden layer, and plot ROC curves with scikit-learn and matplotlib.
Unsupervised Learning with Deep belief Network Final Part5:24
Explore unsupervised learning with a deep belief network, using a TensorFlow 1 compatibility setup as a pre-processing layer for classification that preserves manifold structure and enables a simple linear classifier.

Supervised Learning with Deep Belief Network Implementation Part 114:05
Load a scikit-learn dataset, standardize features, and build a supervised deep belief network with rbm-inspired layers, 16 hidden units and 1 output, using dropout and backprop for glass property classification.
Supervised Learning with Deep Belief Network Implementation Part 37:20
Implement a supervised deep belief network in Python, train with RBM layers and ReLU activation, and evaluate using a classification report for accuracy, precision, and recall.

Requirements

Deep understanding of Artificial Neural Network
Deep understanding of Convolutional Neural Network

Description

Interested in Machine Learning, Deep Learning, and Artificial Intelligence? Then this course is for you!

A software engineer has designed this course. With the experience and knowledge I gained throughout the years, I can share my knowledge and help you learn complex theories, algorithms, and coding libraries.

I will walk you into Deep Belief Networks. There are no courses out there that cover Deep Belief networks. However, Deep Belief Networks are used in many applications such as Image recognition, generation, and clustering, Speech recognition, Video sequences, and Motion capture data. So it is essential to learn and understand Deep Belief Network. With every tutorial, you will develop new skills and improve your understanding of this challenging yet lucrative sub-field of Data Science.

This course is fun and exciting, but at the same time, we dive deep into Deep Belief Networks. Throughout the brand new version of the course, we cover tons of tools and technologies, including:

Google Colab
Deep Belief Network (DBN)
Jupiter Notebook
Artificial Neural Network.
Neuron.
Activation Function.
Keras.
Pandas.
Fine Tuning.
Matplotlib.
Restricted Boltzmann Machines (RBMs)
Contrastive Divergence (CD-k) algorithm
Training DBNs
Bayesian Belief Networks (BBNs)

Moreover, the course is packed with practical exercises based on real-life examples. So not only will you learn the theory, but you will also get some hands-on practice building your models. There are three big projects in this course. These projects are listed below:

MNIST project
Wine project
Movies project.

By the end of the course, you will have a deep understanding of Deep Belief Networks, and you will get a higher chance of getting promoted or a job by knowing Deep belief Networks.

Who this course is for:

Anyone interested in Deep Learning, Machine Learning and Artificial Intelligence
Students who have at least high school knowledge in math and who want to start learning Machine Learning, Deep Learning, and Artificial Intelligence
Any data analysts who want to level up in Machine Learning, Deep Learning and Artificial Intelligence.
Anyone passionate about Artificial Intelligence
Data Scientists who want to take their AI Skills to the next level

Introduction to Deep Belief Network (DBN) with Python 2023

What you'll learn

Explore related topics

Course content

Introduction7 lectures • 34min

RBM Recommender System4 lectures • 48min

Unsupervised Learning with Deep belief Network3 lectures • 29min

Supervised Learning with Deep belief Network2 lectures • 21min

Thank you1 lecture • 1min

Requirements

Description

Who this course is for: