3.95 (53 reviews)
☑ How to build Artificial Neural Networks (ANN) in R
☑ How to build Convolutional Neural Networks (CNN) in R
☑ How to use H20 package in R to solve real world challenges
☑ Read Data Into R Environment From Different Sources
☑ Implement Pre-processing Tasks in R Environment
YOUR COMPLETE GUIDE TO ARTIFICIAL NEURAL NETWORKS & DEEP LEARNING IN R:
This course covers the main aspects of neural networks and deep learning. If you take this course, you can do away with taking other courses or buying books on R based data science.
In this age of big data, companies across the globe use R to sift through the avalanche of information at their disposal. By becoming proficient in neural networks and deep learning in R, you can give your company a competitive edge and boost your career to the next level!
LEARN FROM AN EXPERT DATA SCIENTIST:
My name is Minerva Singh and I am an Oxford University MPhil (Geography and Environment) graduate. I recently finished a PhD at Cambridge University.
I have +5 years of experience in analyzing real life data from different sources using data science related techniques and producing publications for international peer reviewed journals.
Over the course of my research I realized almost all the R data science courses and books out there do not account for the multidimensional nature of the topic .
This course will give you a robust grounding in the main aspects of practical neural networks and deep learning.
Unlike other R instructors, I dig deep into the data science features of R and give you a one-of-a-kind grounding in data science...
You will go all the way from carrying out data reading & cleaning to to finally implementing powerful neural networks and deep learning algorithms and evaluating their performance using R.
Among other things:
You will be introduced to powerful R-based deep learning packages such as h2o and MXNET.
You will be introduced to deep neural networks (DNN), convolution neural networks (CNN) and unsupervised methods.
You will learn how to implement convolutional neural networks (CNN)s on imagery data using the Keras framework
You will learn to apply these frameworks to real life data including credit card fraud data, tumor data, images among others for classification and regression applications.
With this course, you’ll have the keys to the entire R Neural Networks and Deep Learning Kingdom!
Welcome to AI in R course
Welcome To The Course
Install R and RStudio
Data and Code Used in the Course
Install MXnet in R and RStudio
Install Mxnet in R- Written Instructions
What is Keras?
Install Keras in R
Working with Real Data
Read in Data From CSV and Excel Files
Read in Data from Online HTML Tables-Part 1
Read in Data from Online HTML Tables-Part 2
Working with External Data in H2o
More Data Cleaning
Introduction to dplyr for Data Summarizing-Part 1
Introduction to dplyr for Data Summarizing-Part 2
Exploratory Data Analysis(EDA): Basic Visualizations with R
What Are the Most Common Data Types We Will Encounter?
Some Theoretical Foundations
Difference Between Supervised & Unsupervised Learning
Plan of Attack
The Activation Function
How do Neural Networks work?
How do Neural Networks learn?
Stochastic Gradient Descent
Build Artificial Neural Networks (ANN) in R
Neural Network for Binary Classifications
Implement a Multi-Layer Perceptron (MLP) For Supervised Classification
Neural Network for Multiclass Classifications
Neural Network for Image Type Data
Multi-class Classification Using Neural Networks with caret
Implement an ANN with H2o For Multi-Class Supervised Classification
Implement an ANN Based Classification Using MXNet
Implement MLP With Keras
Keras MLP On Real Data
Keras MLP For Regression
Neural Network for Regression
More on Artificial Neural Networks(ANN) - with neuralnet
Implement an ANN Based Regression Using MXNet
Identify Variable Importance in Neural Networks
Build Deep Neural Networks (DNN) in R
Implement a Simple DNN With "neuralnet" for Binary Classifications
Implement a Simple DNN With "deepnet" for Regression
Implement a DNN with H2o For Multi-Class Supervised Classification
Implement a (Less Intensive) DNN with H2o For Supervised Classification
Implement a DNN With Keras
Implement a DNN With Keras
Identify Variable Importance
Implement MXNET via "caret"
Implement a DNN with H2o For Regression
Implement a DNN with Keras For Regression
Implement DNN Regression With Keras (Real Data)
Unsupervised Classification with Deep Learning
Theory Behind Unsupervised Classification
Autoencoders for Unsupervised Learning
Autoencoders for Credit Card Fraud Detection
Use the Autoencoder Model for Anomaly Detection
Autoencoders for Unsupervised Classification
Autoencoders With Keras
Keras Autoencoders on Real Data
Stacked Autoencoder With Keras
Keras For Outlier Detection
Find the Outlier
Outlier Detection For Cancer (With Keras)
Plan of Attack
What are convolutional neural networks?
Step 1 - Convolution Operation
Step 1(b) - ReLU Layer
Step 2 - Pooling
Step 3 - Flattening
Step 4 - Full Connection
Softmax & Cross-Entropy
Practical CNN Implementation in R
Implement a CNN for Multi-Class Supervised Classification
More About Our CNN Model Accuracy
Set Up CNN With Keras
More About CNN With Keras
Implement Keras CNN On Real Images
Some More Explanations
Improve CNN Performance
Working With Textual Data
Basic Pre-Processing of Text Data
Detect Frauds Using Keras Autoencoders on Text Data
Word Embeddings For Classifying Fraud
Word Embeddings For Classifying Fraud-GloVe
The first issue is not letting students know which version of R or R Studio should be installed. Many of the packages included in the training are not compatible with later versions. I also found a few data sets used in the training were not available for download on the SDS site.
just in the welcome capture in the first 7 films allmost all installations instructions went wrong, in codes easy ")" are missing....no...her reaction "you software isn't compatible...but another user found the right command....what she want to teach if the easy installation commands are wrong
Overall, this was a very good course to cover a wide variety of deep learning frameworks in R including MXNet, H2o, and Keras across a variety of different uses cases. The structure is set so that there is a limited amount of theory (which I covered in previous python courses), and more code work which gets into the nuts and bolts of how to actually use these frameworks in R. My version of R (3.6.1) was newer than the one available for MXNet so I skipped those sections; in my work H2o and Keras are far more relevant. The main models focused on both 'toy' data like MNist and some real world data. With some of the Keras models at the end, it would have been great to take it one step further beyond the model accuracy and loss and show the final use - especially in the text/fraud case.
Instructor assumes quite a bit of previous knowledge. I wouldn't recommend to a beginner. MxNet won't install on R 3.6.1.