Practical Deep Learning & Artificial Neural Nets with Python

Name: Practical Deep Learning & Artificial Neural Nets with Python
Rating: 4.7 (7 reviews)

Apply Deep Learning concepts with Python to solve challenging tasks: Detect smiles in your camera app using Neural Nets

Created byPackt Publishing

Last updated 3/2019

English

What you'll learn

Build a solid understanding of common problems can you solve with Deep Learning
Build Deep Neural Networks in the healthcare domain to address applications of deep learning in it
Develop a clear understanding of how Deep Learning tools work and what you need to know to use them in practice
Practical ways in which Deep Learning techniques can be applied to develop solutions for image recognition
Explore face recognition with Deep Learning
Work with dialog generation in Deep Learning
Use different Deep Learning algorithms to solve specific types of problem and learn their strengths and weaknesses,
Save time by learning practical Deep Learning methods that you can immediately apply to real-world problems.

Course content

2 sections • 56 lectures • 6h 27m total length

Course Overview2:17
This video provides an overview of the entire course.
Introduction to Deep Learning and Neural Networks12:01
This video will focus on introduction of neural networks, machine learning and deep learning algorithms.
• Understand neural networks
• Focus on machine learning and deep learning aspects
• Use these concepts for upcoming sections
Building Neural Network8:14
This video will teach how neural network always focusses on creating training and testing data for better evaluation.
• Understand the problem statement
• Focus on the attributes which needs to be changed
• Generate output for first neural network evaluation
Evaluating the Neural Network8:08
This video focus on training and testing data and how it plays important role in implementing deep learning algorithms.
• Evaluate the data attributes
• Focus on training and testing data
• Check on the output
Ohio Clinic Data Set1:43
This video will focus on the problem statement with reference to clinic/ hospital dataset. This hospital includes various records and sections which needs to be evaluated.
• Understand the attributes
• Focus on the type of diseases and sections
• Analyze the data attributes
Analyze and Explore Your Data10:10
This video will import the necessary modules and start evaluating them.
• Launch Jupyter notebook, upload the csv file of hospital
• Evaluate the features when necessary modules are imported
• Correct and analyze the data in systematic manner
Feature Exploration of Our Dataset7:07
This video will evaluate all features to get the visualization.
• Focus on the attributes to be created
• Evaluate and convert them in that specific order
• Create the respective chart for same
Performing Data Analysis7:26
This video will teach the use of pandas for data and time conversion with data analysis.
• Convert the date in specified format
• Append the converted data in specified manner
• Analyze the output
Introduction to Image Recognition1:25
This video will focus on the problem statement, the need for creating image recognition module.
• Understand image recognition concept
• Focus on the problem statement
• Focus on evaluation
Environmental Setup5:44
This video will focus on data exploration and the environmental setup of OpenCV library which is needed for image, facial and body recognition.
• Explore the data
• Understand the installation of OpenCV from documentation
• Evaluate and explore the possibilities
Using Spyder IDE8:47
This video will help you create a pattern of CSV for same.
• The CSV will include list of images which needs to be encoded and training
• Evaluate and convert them in that specific order
• Create the respective model for same
Encode the Image6:44
This video, we will use harcascade xml for encode the image pattern.
• Understanding the importance of xml
• Encode the parameters
• Understand testing functionality and output
Understanding Testing Functionality and Output7:01
This video will focus on the evaluation and create a classification model for image recognition and analyzing the output.
• Create a training model
• Focus on the image and visualize the facial parameter
• Analyze the output
Introduction to Face Recognition1:59
This video will focus on the problem evaluation and understand why facial recognition is needed.
• Understand the need of facial recognition of patients
• Focus where images will be stored
• Analyze the data attributes
Problem Statement1:29
This video will focus on the problem statement and understand how it is useful from security point of view.
• Understand the problem
• Analyze the enhancement of security feature
Face Recognition File and Output11:41
This video will focus on the output.
• Execute the code
• Code will set on the web cam and capture the images
• All the images will be saved in datasets folder
Introduction to Keras1:24
This video, will make you understand the problem statement with the mentioned dataset. In this video, we will focus on critical causalities and create a time series module of same.
• Understand the dataset
• Focus on the number of records
• Analyze the data attributes and environmental setup of Keras
Feedforward Neural Network3:00
This video will focus on working module of LSTM.
• Understand the network architecture
• LSTM input and output structure
• Understand how the model can be evaluated
Representing Simple FeedForward Neural Network Using Keras5:58
This video will teaches us how to create LSTM model with base of recurrent neural network for classification of cats and dogs images.
• Focus on the model creation
• Encode the parameters
• Analyze the output
Scaling Input Images6:16
This video will focus on the output.
• Execute the code
• The classifier will create a separate folder structure of input and output data
• All the images will be saved in datasets folder
Introduction to LSTM1:40
This video will focus on the problem statement with the mentioned dataset. In this video, we will focus on critical causalities and create a time series module of same.
• Understand the dataset
• Focus on the number of records
• Analyze the data attributes and environmental setup of Keras
LSTM Architecture1:45
This video will focus on working module of LSTM.
• Understand the network architecture
• LSTM input and output structure
• Understand how the model can be evaluated. Implement the model with LSTM structure
How LSTM Network Works5:54
This video teaches us how to create an LSTM model with base of recurrent neural network for document characterization.
• Focus on the model creation
• Encode the parameters
• Analyze the output
Fitting Neural Network and Output4:33
This video will focus on the output.
• Execute the code
• The classifier will create a separate folder structure of input and output data
• The list of data will be evaluated in respective chart
Introduction to Text Summarization1:30
This video will focus on the problem evaluation and understand how junks of data such as case files be converted in specific format
• Understand the need of text summarization
• Focus on methodology
• Analyze the data attributes.
Understanding the Problem Statement2:45
This video will focus on the problem statement with respect to text summarization
• Focus on the problem statement
• Analyze the manner in which case files are maintained.
• Focus on model creation
Training and Testing Data3:47
This video we will focus on creating testing and training data for better evaluation
• Understanding the importance of both
• Encode the parameters
• Focusing on preparation data
Preparation Data4:10
This video will focus on the output
• Execute the code
• The separate folders will be created accordingly
• Text summarization procedure will be followed.
Introduction to Encode-Decode Model2:12
This video will focus on dialog generation with encode and decode model
• Understand the need of encode and decode model
• Focus on methodology
• Analyze the data attributes.
Implementing Decoder and Encoder5:11
This video will focus on the problem statement with respect to dialog generation and evaluate the same
• Focus on the problem statement
• Analyze the manner in code prediction of data will be maintained
• Focus on model creation
Defining the Module and Output5:22
This video will focus on the output and creation of accuracy rate
• Understanding the output generation
• Check out the accuracy rate
• Analyze the output.
Test your knowledge

The Course Overview5:18
This video provides an overview of the entire course.
What Types of Problems Can You Solve Using Deep Learning?4:34
The goal of this video is to help you understand why you should learn deep learning and in what cases it's beneficial to use it.
Define what deep learning is and why it's different from other methods
Present the general types of problems that deep learning is good at solving
Give specific examples of problems that deep learning helps to solve
Installing Essential DL Tools8:16
The goal of this video is to show you how you can quickly install the necessary tools that we will be using throughout the course.
Introduce the package manager that we will be using to install all the required tools, and learn why you should use it
Give a quick description of tools we will be using and in the course and installing in the video
Show how to quickly install all the necessary tools step by step
Based on Past Data, Predicting the Number of Airline Passengers2:37
An overview of the process of solving a time series prediction problem using deep learning methods.
The first step is to phrase our problem in the correct way and prepare data for working with a neural network.
Then we need to build our neural network in Python.
The last step is to train our model with our data and tweak it for best performance. Then we can just use it make a prediction.
Getting and Preparing Airline Data8:25
Here we will be downloading and preparing our airline data to work without a neural network.
Download data from the right place in the right format
Phrase our problem as a regression problem and convert our data into the right format
Verify that all of the data is ready for our neural network
Building Your Multilayer Perceptron Model8:15
In this Section, we will build our first neural network in Python for making predictions.
Learn what MLP is
Build MLP in Python
Understand MLP’s layers
Training and Testing Your Model23:55
Here you will learn how to train and test your model for best performance.
Understand the hyperparameters that influence training
Train your model and watch the important metrics
Tweak the parameters according to the results you get in training
Making Predictions and What's Next?8:47
Let’s make predictions using our model and discover what’s next.
Load the saved model
Choose the right value for prediction
Make the prediction using our model
End Goal – Label a Given Tweet (Short Text) as Negative or Positive2:26
We’re starting with a quick overview of what you will learn in this Section.
Define the project’s goals, including our end goal
Define the steps to get there
Explain what you will learn in each step
Dataset Overview5:14
Next, we will have a look at our dataset to understand how we can use it.
Which dataset we will be using and why
How our dataset is organized
Understanding the dataset format
Preparing Data for Sentiment Analysis15:00
Learn how to prepare text to work with text classification using deep learning methods.
Choose the right dataset and load it into memory
Clean up the data
Encode the data
What Are Word Embeddings and Why They Are Important When Working with CNNs?7:41
Understand how we can use a CNN network designed to work with images to work with text.
The main idea behind word embeddings
How to turn text into word embeddings
How we can use word embeddings with our CNN network
Building Your CNN Model for Text Classification11:54
In this video, you will learn how to build a CNN network that can perform sentiment analysis.
Learn how to use the Embedding layer and what word embeddings are
Discover the main parts of every CNN model—Convolution Layer and Pooling layer—and how they work together
Learn how to define the last layer for the classification problem
Training and Testing Your Model12:17
Next, you will learn how to train and test your text classification model.
The key parameters to set up before training your model
Understand the main metrics that will help you pick up the best model
Two ways to get better results with your text classification model
Detecting Mean Tweets Using Your Model and What’s Next?15:55
In this last video, you will learn how to use your model to judge a tweet and learn ways to move forward.
Load the model and tokenizer saved during training
Read the tweets for analysis and encode them using the tokenizer
Use two methods to judge whether a tweet is negative or not
Detect Whether an Image Contains a Smile with High Accuracy1:51
Get a quick overview of this project and see the steps to build it.
Clarify our goals for this project
Go through the steps to complete the project
Briefly describe what we will be doing in each step
Getting and Preparing Data for Smile Detection13:36
In this video, we will be preparing our input images to work with CNN.
Download images and put them in the right folders
Convert each into a two-dimensional array and optimize them for best results
Split the dataset into training and testing parts
Building Your CNN Model for Smile Detection16:52
You will learn the basics of building a CNN for image classification.
Learn how to set up an input convolutional layer and output layer
Understanding intuitively how CNNs work
Learn the role of the rest of the layers in a CNN, choosing the right loss function and optimizer to detect multiple classes
Training and Testing Your Model7:12
In this video, you will find the best parameters to train your CNN model.
Quickly train a model with different parameters of number of epochs and batch size
Recognize which set of parameters is the best
Train the final model with the best parameters
Detecting Smiles with Your Model and What’s Next?15:59
Here, we will be using our model to detect smiles in a totally new dataset and discuss what we can do to improve the accuracy of our model.
Get familiar with our new dataset, run our prediction script, and understand the output metrics
Examine the photos that were classified incorrectly
Discuss some ways to improve the accuracy of our model
Predict the Closing Stock Price of a Given Company for the Next Day1:29
Get a high-level overview of the project.
Clearly define the project’s goals
Break down the project into steps
Get a quick overview of each step
Getting and Preparing Stock Prices Data9:56
Learn how to get the free stock prices data and prepare it for forecasting using LSTM.
Download the historical stock prices data for a given company
Remove the unnecessary columns from the dataset
Convert the dataset into a supervised learning problem
Building Your LSTM Model for Price Prediction7:16
Create and compile an LSTM model for time series forecasting.
Create the main model container
Add and configure an LSTM layer
Add the last Dense layer for price prediction
Training and Testing Your Model4:50
Find the optimal training parameters for our LSTM model.
Train the model with initial parameters and chart the training metrics
Interpret the training chart
Train the model with optimal parameters
Detecting Closing Stock Price with Your Model and What’s Next?10:53
Learn how to use our LSTM model to predict stock prices using historical stock metrics. Explore future improvements.
Prepare historical metrics to be used for prediction
Reshape the data so that it can be used for prediction using LSTM
Make a stock price prediction, compare it with the actual price, and calculate the prediction error
Test your knowledge

Requirements

To pick up this course, you need to have Python programming skills. Developers, analysts, and data scientists who have a basic Machine Learning knowledge and want to now explore the possibilities of Deep Learning will feel perfectly comfortable in understanding the topics presented in this Course.

Description

Video Learning Path Overview

A Learning Path is a specially tailored course that brings together two or more different topics that lead you to achieve an end goal. Much thought goes into the selection of the assets for a Learning Path, and this is done through a complete understanding of the requirements to achieve a goal.

Deep learning is the next step to a more advanced implementation of Machine Learning. Deep Learning allows you to solve problems where traditional Machine Learning methods might perform poorly: detecting and extracting objects from images, extracting meaning from text, and predicting outcomes based on complex dependencies, to name a few.

In this practical Learning Path, you will build Deep Learning applications with real-world datasets and Python. Beginning with a step by step approach, right from building your neural nets to reinforcement learning and working with different Deep Learning applications such as computer Vision and voice and image recognition, this course will be your guide in getting started with Deep Learning concepts.

Moving further with simple and practical solutions provided, we will cover a whole range of practical, real-world projects that will help customers learn how to implement their skills to solve everyday problems.

By the end of the course, you’ll apply Deep Learning concepts and use Python to solve challenging tasks with real-world datasets.

Key Features

Get started with Deep Learning and build complex models layer by layer, with increasing complexity, in no time.
A hands-on guide covering common as well as not-so-common problems in deep learning using Python.
Explore the practical essence of Deep Learning in a relatively short amount of time by working on practical, real-world use cases.

Author Bios

Radhika Datar has more than 6 years' experience in Software Development and Content Writing. She is well versed with frameworks such as Python, PHP, and Java and regularly provides training on them. She has been working with Educba and Eduonix as a Training Consultant since June 2016 and has been an Academic writer with TutorialsPoint since Sept 2015.
Jakub Konczyk has enjoyed and done programming professionally since 1995. He is a Python and Django expert and has been involved in building complex systems since 2006. He loves to simplify and teach programming subjects and share it with others. He first discovered Machine Learning when he was trying to predict the real estate prices in one of the early stage start-ups he was involved in. He failed miserably but then discovered a much more practical way to learn Machine Learning that he shares in this course.

Who this course is for:

Data Science Professionals, Machine Learning enthusiasts, Developers, Analysts, who would like to gain practical hands-on experience to their Deep Learning problems and build Deep-Learning applications with real-world datasets in Python, will find this course useful.

Practical Deep Learning & Artificial Neural Nets with Python

What you'll learn

Explore related topics

Course content

Hands-On Python Deep Learning31 lectures • 2hr 37min

Real-World Python Deep Learning Projects25 lectures • 3hr 50min

Requirements

Description

Who this course is for: