Search icon Close icon
Search icon CANCEL
Subscription
0
Cart icon
Your Cart (0 item)
Close icon
You have no products in your basket yet
Checkout

Change country

Modal Close icon
Country selected Country selected
country flag United States
Country selected
country flag Great Britain
Country selected
country flag India
Country selected
country flag Germany
Country selected
country flag France
Country selected
country flag Canada
Country selected
country flag Russia
Country selected
country flag Spain
Country selected
country flag Brazil
Country selected
country flag Australia
Country selected
country flag Singapore
Country selected
country flag Hungary
Country selected
country flag Ukraine
Country selected
country flag Luxembourg
Country selected
country flag Estonia
Country selected
country flag Lithuania
Country selected
country flag South Korea
Country selected
country flag Turkey
Country selected
country flag Switzerland
Country selected
country flag Colombia
Country selected
country flag Taiwan
Country selected
country flag Chile
Country selected
country flag Norway
Country selected
country flag Ecuador
Country selected
country flag Indonesia
Country selected
country flag New Zealand
Country selected
country flag Cyprus
Country selected
country flag Denmark
Country selected
country flag Finland
Country selected
country flag Poland
Country selected
country flag Malta
Country selected
country flag Czechia
Country selected
country flag Austria
Country selected
country flag Sweden
Country selected
country flag Italy
Country selected
country flag Egypt
Country selected
country flag Belgium
Country selected
country flag Portugal
Country selected
country flag Slovenia
Country selected
country flag Ireland
Country selected
country flag Romania
Country selected
country flag Greece
Country selected
country flag Argentina
Country selected
country flag Netherlands
Country selected
country flag Bulgaria
Country selected
country flag Latvia
Country selected
country flag South Africa
Country selected
country flag Malaysia
Country selected
country flag Japan
Country selected
country flag Slovakia
Country selected
country flag Philippines
Country selected
country flag Mexico
Country selected
country flag Thailand
Country selected
Arrow left icon
Explore Products
Best Sellers
New Releases
Books
Videos
Audiobooks
Learning Hub
Free Learning
Arrow right icon
Home > Data > Deep Learning > Deep Learning with Theano
Deep Learning with Theano
Deep Learning with Theano

Deep Learning with Theano: Perform large-scale numerical and scientific computations efficiently

Arrow left icon
Profile Icon Bourez
Arrow right icon
$48.99
Full star icon Full star icon Full star icon Half star icon Empty star icon 3.7 (3 Ratings)
Paperback Jul 2017 300 pages 1st Edition
eBook
$9.99 $39.99
Paperback
$48.99
Subscription
Free Trial
Renews at $19.99p/m
Arrow left icon
Profile Icon Bourez
Arrow right icon
$48.99
Full star icon Full star icon Full star icon Half star icon Empty star icon 3.7 (3 Ratings)
Paperback Jul 2017 300 pages 1st Edition
eBook
$9.99 $39.99
Paperback
$48.99
Subscription
Free Trial
Renews at $19.99p/m
eBook
$9.99 $39.99
Paperback
$48.99
Subscription
Free Trial
Renews at $19.99p/m

What do you get with Print?

Product feature icon Instant access to your digital eBook copy whilst your Print order is Shipped
Product feature icon Paperback book shipped to your preferred address
Product feature icon Download this book in EPUB and PDF formats
Product feature icon Access this title in our online reader with advanced features
Product feature icon DRM FREE - Read whenever, wherever and however you want
OR

Contact Details

Modal Close icon
Payment Processing...
tick Completed

Shipping Address

Billing Address

Shipping Methods
Table of content icon View table of contents Preview book icon Preview Book

Deep Learning with Theano

Chapter 2. Classifying Handwritten Digits with a Feedforward Network

The first chapter presented Theano as a compute engine, with its different functions and specificities. With this knowledge, we'll go through an example and introduce some of the main concepts of deep learning, building three neural networks and training them on the problem of handwritten digit classification.

Deep learning is a field of machine learning in which layers of modules are stacked on top of each of other: this chapter introduces a simple single-linear-layer model, then adds a second layer on top of it to create a multi-layer perceptron (MLP), and last uses multiple convolutional layers to create a Convolutional Neural Network (CNN).

In the meantime, this chapter recaps the basic machine learning concepts, such as overfitting, validation, and loss analysis, for those who are not familiar with data science:

  • Small image classification
  • Handwritten digit recognition challenge
  • Layer design to build a neural...

The MNIST dataset

The Modified National Institute of Standards and Technology (MNIST) dataset is a very well-known dataset of handwritten digits {0,1,2,3,4,5,6,7,8,9} used to train and test classification models.

A classification model is a model that predicts the probabilities of observing a class, given an input.

Training is the task of learning the parameters to fit the model to the data as well as we can so that for any input image, the correct label is predicted. For this training task, the MNIST dataset contains 60,000 images with a target label (a number between 0 and 9) for each example.

To validate that the training is efficient and to decide when to stop the training, we usually split the training dataset into two datasets: 80% to 90% of the images are used for training, while the remaining 10-20% of images will not be presented to the algorithm for training but to validate that the model generalizes well on unobserved data.

There is a separate dataset that the algorithm should never...

Structure of a training program

The structure of a training program always consists of the following steps:

  1. Set the script environment: Such as package imports, the use of the GPU, and so on.
  2. Load data: A data loader class to access the data during training, usually in a random order to avoid too many similar examples of the same class, but sometimes in a precise order, for example, in the case of curriculum learning with simple examples first and complex ones last.
  3. Preprocess the data: A set of transformations, such as swapping dimensions on images, adding blur or noise. It is very common to add some data augmentation transformations, such as random crop, scale, brightness, or contrast jittering to get more examples than the original ones, and reduce the risk of overfitting on data. If the number of free parameters in the model is too important with respect to the training dataset size, the model might learn from the available examples. Also, if the dataset is too small and too many iterations...

Classification loss function

The loss function is an objective function to minimize during training to get the best model. Many different loss functions exist.

In a classification problem, where the target is to predict the correct class among k classes, cross-entropy is commonly used as it measures the difference between the real probability distribution, q, and the predicted one, p, for each class:

Classification loss function

Here, i is the index of the sample in the dataset, n is the number of samples in the dataset, and k is the number of classes.

While the real probability Classification loss function of each class is unknown, it can simply be approximated in practice by the empirical distribution, that is, randomly drawing a sample out of the dataset in the dataset order. The same way, the cross-entropy of any predicted probability, p, can be approximated by the empirical cross-entropy:

Classification loss function

Here, Classification loss function is the probability estimated by the model for the correct class of example Classification loss function.

Accuracy and cross-entropy both evolve in the same direction but measure...

Single-layer linear model

The simplest model is the linear model, where for each class c, the output is a linear combination of the input values:

Single-layer linear model

This output is unbounded.

To get a probability distribution, pi, that sums to 1, the output of the linear model is passed into a softmax function:

Single-layer linear model

Hence, the estimated probability of class c for an input x is rewritten with vectors:

Single-layer linear model

Translated in Python with:

batch_size = 600
n_in = 28 * 28
n_out = 10

x = T.matrix('x')
y = T.ivector('y')
W = theano.shared(
            value=numpy.zeros(
                (n_in, n_out),
                dtype=theano.config.floatX
            ),
            name='W',
            borrow=True
        )
b = theano.shared(
    value=numpy.zeros(
        (n_out,),
        dtype=theano.config.floatX
    ),
    name='b',
    borrow=True
)
model = T.nnet.softmax(T.dot(x, W) + b)

The prediction for a given input is given by the most probable class (maximum probability):

y_pred = T.argmax(model...

Cost function and errors

The cost function given the predicted probabilities by the model is as follows:

cost = -T.mean(T.log(model)[T.arange(y.shape[0]), y])

The error is the number of predictions that are different from the true class, averaged by the total number of values, which can be written as a mean:

error = T.mean(T.neq(y_pred, y))

On the contrary, accuracy corresponds to the number of correct predictions divided by the total number of predictions. The sum of error and accuracy is one.

For other types of problems, here are a few other loss functions and implementations:

Categorical cross entropy

An equivalent implementation of ours

 

T.nnet.categorical_crossentropy(model, y_true).mean()

Binary cross entropy

For the case when output can take only two values {0,1}

Typically used after a sigmoid activation predicting the probability, p

T.nnet.binary_crossentropy(model, y_true).mean()

Mean squared error

L2 norm for regression problems

T.sqr(model – y_true).mean()

Mean absolute...

The MNIST dataset

The Modified National Institute of Standards and Technology (MNIST) dataset is a very well-known dataset of handwritten digits {0,1,2,3,4,5,6,7,8,9} used to train and test classification models.

A classification model is a model that predicts the probabilities of observing a class, given an input.

Training is the task of learning the parameters to fit the model to the data as well as we can so that for any input image, the correct label is predicted. For this training task, the MNIST dataset contains 60,000 images with a target label (a number between 0 and 9) for each example.

To validate that the training is efficient and to decide when to stop the training, we usually split the training dataset into two datasets: 80% to 90% of the images are used for training, while the remaining 10-20% of images will not be presented to the algorithm for training but to validate that the model generalizes well on unobserved data.

There is a separate dataset that the algorithm should never...

Structure of a training program

The structure of a training program always consists of the following steps:

  1. Set the script environment: Such as package imports, the use of the GPU, and so on.

  2. Load data: A data loader class to access the data during training, usually in a random order to avoid too many similar examples of the same class, but sometimes in a precise order, for example, in the case of curriculum learning with simple examples first and complex ones last.

  3. Preprocess the data: A set of transformations, such as swapping dimensions on images, adding blur or noise. It is very common to add some data augmentation transformations, such as random crop, scale, brightness, or contrast jittering to get more examples than the original ones, and reduce the risk of overfitting on data. If the number of free parameters in the model is too important with respect to the training dataset size, the model might learn from the available examples. Also, if the dataset is too small and too many iterations...

Classification loss function

The loss function is an objective function to minimize during training to get the best model. Many different loss functions exist.

In a classification problem, where the target is to predict the correct class among k classes, cross-entropy is commonly used as it measures the difference between the real probability distribution, q, and the predicted one, p, for each class:

Here, i is the index of the sample in the dataset, n is the number of samples in the dataset, and k is the number of classes.

While the real probability of each class is unknown, it can simply be approximated in practice by the empirical distribution, that is, randomly drawing a sample out of the dataset in the dataset order. The same way, the cross-entropy of any predicted probability, p, can be approximated by the empirical cross-entropy:

Here, is the probability estimated by the model for the correct class of example .

Accuracy and cross-entropy both evolve in the same direction but measure...

Single-layer linear model

The simplest model is the linear model, where for each class c, the output is a linear combination of the input values:

This output is unbounded.

To get a probability distribution, pi, that sums to 1, the output of the linear model is passed into a softmax function:

Hence, the estimated probability of class c for an input x is rewritten with vectors:

Translated in Python with:

batch_size = 600
n_in = 28 * 28
n_out = 10

x = T.matrix('x')
y = T.ivector('y')
W = theano.shared(
            value=numpy.zeros(
                (n_in, n_out),
                dtype=theano.config.floatX
            ),
            name='W',
            borrow=True
        )
b = theano.shared(
    value=numpy.zeros(
        (n_out,),
        dtype=theano.config.floatX
    ),
    name='b',
    borrow=True
)
model = T.nnet.softmax(T.dot(x, W) + b)

The prediction for a given input is given by the most probable class (maximum probability):

y_pred = T.argmax(model, axis=1)

In this model with a single linear...

Cost function and errors

The cost function given the predicted probabilities by the model is as follows:

cost = -T.mean(T.log(model)[T.arange(y.shape[0]), y])

The error is the number of predictions that are different from the true class, averaged by the total number of values, which can be written as a mean:

error = T.mean(T.neq(y_pred, y))

On the contrary, accuracy corresponds to the number of correct predictions divided by the total number of predictions. The sum of error and accuracy is one.

For other types of problems, here are a few other loss functions and implementations:

Categorical cross entropy

An equivalent implementation of ours

 

T.nnet.categorical_crossentropy(model, y_true).mean()

Binary cross entropy

For the case when output can take only two values {0,1}

Typically used after a sigmoid activation predicting the probability, p

T.nnet.binary_crossentropy(model, y_true).mean()

Mean squared error

L2 norm for regression problems

T.sqr(model – y_true).mean()

Mean absolute...

Backpropagation and stochastic gradient descent

Backpropagation, or the backward propagation of errors, is the most commonly used supervised learning algorithm for adapting the connection weights.

Considering the error or the cost as a function of the weights W and b, a local minimum of the cost function can be approached with a gradient descent, which consists of changing weights along the negative error gradient:

Here, is the learning rate, a positive constant defining the speed of a descent.

The following compiled function updates the variables after each feedforward run:

g_W = T.grad(cost=cost, wrt=W)
g_b = T.grad(cost=cost, wrt=b)

learning_rate=0.13
index = T.lscalar()

train_model = theano.function(
    inputs=[index],
    outputs=[cost,error],
    updates=[(W, W - learning_rate * g_W),(b, b - learning_rate * g_b)],
    givens={
        x: train_set_x[index * batch_size: (index + 1) * batch_size],
        y: train_set_y[index * batch_size: (index + 1) * batch_size]
    }
)

The input variable...

Multiple layer model

A multi-layer perceptron (MLP) is a feedforward net with multiple layers. A second linear layer, named hidden layer, is added to the previous example:

Having two linear layers following each other is equivalent to having a single linear layer.

With a non-linear function or non-linearity or transfer function between the linearities, the model does not simplify into a linear one any more, and represents more possible functions in order to capture more complex patterns in the data:

Activation functions helps saturating (ON-OFF) and reproduces the biological neuron activations.

The Rectified Linear Unit (ReLU) graph is given as follows:

(x + T.abs_(x)) / 2.0

The Leaky Rectifier Linear Unit (Leaky ReLU) graph is given as follows:

( (1 + leak) * x + (1 – leak) * T.abs_(x) ) / 2.0

Here, leak is a parameter that defines the slope in the negative values. In leaky rectifiers, this parameter is fixed.

The activation named PReLU considers the leak parameter to be learned.

More generally...

Convolutions and max layers

A great improvement in image classification has been achieved with the invention of the convolutional layers on the MNIST database:

While previous fully-connected layers perform a computation with all input values (pixels in the case of an image) of the input, a 2D convolution layer will consider only a small patch or window or receptive field of NxN pixels of the 2D input image for each output unit. The dimensions of the patch are named kernel dimensions, N is the kernel size, and the coefficients/parameters are the kernel.

At each position of the input image, the kernel produces a scalar, and all position values will lead to a matrix (2D tensor) called a feature map. Convolving the kernel on the input image as a sliding window creates a new output image. The stride of the kernel defines the number of pixels to shift the patch/window over the image: with a stride of 2, the convolution with the kernel is computed every 2 pixels.

For example, on a 224 x 224 input...

Training

In order to get a good measure of how the model behaves on data that's unseen during training, the validation dataset is used to compute a validation loss and accuracy during training.

The validation dataset enables us to choose the best model, while the test dataset is only used at the end to get the final test accuracy/error of the model. The training, test, and validation datasets are discrete datasets, with no common examples. The validation dataset is usually 10 times smaller than the test dataset to slow the training process as little as possible. The test dataset is usually around 10-20% of the training dataset. Both the training and validation datasets are part of the training program, since the first one is used to learn, and the second is used to select the best model on unseen data at training time.

The test dataset is completely outside the training process and is used to get the accuracy of the produced model, resulting from training and model selection.

If the model overfits...

Left arrow icon

Page 1 of 15

Right arrow icon
Download code icon Download Code

Key benefits

  • Learn Theano basics and evaluate your mathematical expressions faster and in an efficient manner
  • Learn the design patterns of deep neural architectures to build efficient and powerful networks on your datasets
  • Apply your knowledge to concrete fields such as image classification, object detection, chatbots, machine translation, reinforcement agents, or generative models.

Description

This book offers a complete overview of Deep Learning with Theano, a Python-based library that makes optimizing numerical expressions and deep learning models easy on CPU or GPU. The book provides some practical code examples that help the beginner understand how easy it is to build complex neural networks, while more experimented data scientists will appreciate the reach of the book, addressing supervised and unsupervised learning, generative models, reinforcement learning in the fields of image recognition, natural language processing, or game strategy. The book also discusses image recognition tasks that range from simple digit recognition, image classification, object localization, image segmentation, to image captioning. Natural language processing examples include text generation, chatbots, machine translation, and question answering. The last example deals with generating random data that looks real and solving games such as in the Open-AI gym. At the end, this book sums up the best -performing nets for each task. While early research results were based on deep stacks of neural layers, in particular, convolutional layers, the book presents the principles that improved the efficiency of these architectures, in order to help the reader build new custom nets.

Who is this book for?

This book is indented to provide a full overview of deep learning. From the beginner in deep learning and artificial intelligence, to the data scientist who wants to become familiar with Theano and its supporting libraries, or have an extended understanding of deep neural nets. Some basic skills in Python programming and computer science will help, as well as skills in elementary algebra and calculus.

What you will learn

  • Get familiar with Theano and deep learning
  • Provide examples in supervised, unsupervised, generative, or reinforcement learning.
  • Discover the main principles for designing efficient deep learning nets: convolutions, residual connections, and recurrent connections.
  • Use Theano on real-world computer vision datasets, such as for digit classification and image classification.
  • Extend the use of Theano to natural language processing tasks, for chatbots or machine translation
  • Cover artificial intelligence-driven strategies to enable a robot to solve games or learn from an environment
  • Generate synthetic data that looks real with generative modeling
  • Become familiar with Lasagne and Keras, two frameworks built on top of Theano
Estimated delivery fee Deliver to United States

Economy delivery 10 - 13 business days

Free $6.95

Premium delivery 6 - 9 business days

$21.95
(Includes tracking information)

Product Details

Country selected
Publication date, Length, Edition, Language, ISBN-13
Publication date : Jul 31, 2017
Length: 300 pages
Edition : 1st
Language : English
ISBN-13 : 9781786465825
Vendor :
Google
Category :
Data
Concepts :
Deep Learning
Tools :
Theano

What do you get with Print?

Product feature icon Instant access to your digital eBook copy whilst your Print order is Shipped
Product feature icon Paperback book shipped to your preferred address
Product feature icon Download this book in EPUB and PDF formats
Product feature icon Access this title in our online reader with advanced features
Product feature icon DRM FREE - Read whenever, wherever and however you want
OR

Contact Details

Modal Close icon
Payment Processing...
tick Completed

Shipping Address

Billing Address

Shipping Methods
Estimated delivery fee Deliver to United States

Economy delivery 10 - 13 business days

Free $6.95

Premium delivery 6 - 9 business days

$21.95
(Includes tracking information)

Product Details

Publication date : Jul 31, 2017
Length: 300 pages
Edition : 1st
Language : English
ISBN-13 : 9781786465825
Vendor :
Google
Category :
Data
Concepts :
Deep Learning
Tools :
Theano

Packt Subscriptions

See our plans and pricing
Modal Close icon
$19.99 billed monthly
Feature tick icon Unlimited access to Packt's library of 7,000+ practical books and videos
Feature tick icon Constantly refreshed with 50+ new titles a month
Feature tick icon Exclusive Early access to books as they're written
Feature tick icon Solve problems while you work with advanced search and reference features
Feature tick icon Offline reading on the mobile app
Feature tick icon Simple pricing, no contract
START FREE TRIAL
BUY NOW
$199.99 billed annually
Feature tick icon Unlimited access to Packt's library of 7,000+ practical books and videos
Feature tick icon Constantly refreshed with 50+ new titles a month
Feature tick icon Exclusive Early access to books as they're written
Feature tick icon Solve problems while you work with advanced search and reference features
Feature tick icon Offline reading on the mobile app
Feature tick icon Choose a DRM-free eBook or Video every month to keep
Feature tick icon PLUS own as many other DRM-free eBooks or Videos as you like for just $5 each
Feature tick icon Exclusive print discounts
START FREE TRIAL
BUY NOW
$279.99 billed in 18 months
Feature tick icon Unlimited access to Packt's library of 7,000+ practical books and videos
Feature tick icon Constantly refreshed with 50+ new titles a month
Feature tick icon Exclusive Early access to books as they're written
Feature tick icon Solve problems while you work with advanced search and reference features
Feature tick icon Offline reading on the mobile app
Feature tick icon Choose a DRM-free eBook or Video every month to keep
Feature tick icon PLUS own as many other DRM-free eBooks or Videos as you like for just $5 each
Feature tick icon Exclusive print discounts
START FREE TRIAL
BUY NOW

Frequently bought together

Deep Learning with Theano
Deep Learning with Theano
Read more
Jul 2017 300 pages
Full star icon 3.7 (3)
eBook
eBook
  • eBook eBook $9.99
  • Paperback Paperback $48.99
$9.99 $39.99
$48.99
Python Deep Learning
Python Deep Learning
Read more
Apr 2017 406 pages
Full star icon 4.1 (10)
eBook
eBook
  • eBook eBook $9.99
  • Paperback Paperback $60.99
$9.99 $47.99
$60.99
Deep Learning with TensorFlow
Deep Learning with TensorFlow
Read more
Apr 2017 320 pages
Full star icon 2 (10)
eBook
eBook
  • eBook eBook $9.99
  • Paperback Paperback $54.99
$9.99 $43.99
$54.99
Stars icon
Total $ 164.97
Deep Learning with Theano
$48.99
Python Deep Learning
$60.99
Deep Learning with TensorFlow
$54.99
Total $ 164.97 Stars icon
Banner background image

Table of Contents

14 Chapters
1. Theano Basics Chevron down icon Chevron up icon
1. Theano Basics
The need for tensors
Installing and loading Theano
Tensors
Graphs and symbolic computing
Operations on tensors
Memory and variables
Functions and automatic differentiation
Loops in symbolic computing
Configuration, profiling and debugging
Summary
2. Classifying Handwritten Digits with a Feedforward Network Chevron down icon Chevron up icon
2. Classifying Handwritten Digits with a Feedforward Network
The MNIST dataset
Structure of a training program
Classification loss function
Single-layer linear model
Cost function and errors
Backpropagation and stochastic gradient descent
Multiple layer model
Convolutions and max layers
Training
Dropout
Inference
Optimization and other update rules
Related articles
Summary
3. Encoding Word into Vector Chevron down icon Chevron up icon
3. Encoding Word into Vector
Encoding and embedding
Dataset
Continuous Bag of Words model
Training the model
Visualizing the learned embeddings
Evaluating embeddings – analogical reasoning
Evaluating embeddings – quantitative analysis
Application of word embeddings
Weight tying
Further reading
Summary
4. Generating Text with a Recurrent Neural Net Chevron down icon Chevron up icon
4. Generating Text with a Recurrent Neural Net
Need for RNN
A dataset for natural language
Simple recurrent network
Metrics for natural language performance
Training loss comparison
Example of predictions
Applications of RNN
Related articles
Summary
5. Analyzing Sentiment with a Bidirectional LSTM Chevron down icon Chevron up icon
5. Analyzing Sentiment with a Bidirectional LSTM
Installing and configuring Keras
Preprocessing text data
Designing the architecture for the model
Compiling and training the model
Evaluating the model
Saving and loading the model
Running the example
Further reading
Summary
6. Locating with Spatial Transformer Networks Chevron down icon Chevron up icon
6. Locating with Spatial Transformer Networks
MNIST CNN model with Lasagne
A localization network
Unsupervised learning with co-localization
Region-based localization networks
Further reading
Summary
7. Classifying Images with Residual Networks Chevron down icon Chevron up icon
7. Classifying Images with Residual Networks
Natural image datasets
Residual connections
Stochastic depth
Dense connections
Multi-GPU
Data augmentation
Further reading
Summary
8. Translating and Explaining with Encoding – decoding Networks Chevron down icon Chevron up icon
8. Translating and Explaining with Encoding – decoding Networks
Sequence-to-sequence networks for natural language processing
Seq2seq for translation
Seq2seq for chatbots
Improving efficiency of sequence-to-sequence network
Deconvolutions for images
Multimodal deep learning
Further reading
Summary
9. Selecting Relevant Inputs or Memories with the Mechanism of Attention Chevron down icon Chevron up icon
9. Selecting Relevant Inputs or Memories with the Mechanism of Attention
Differentiable mechanism of attention
Store and retrieve information in Neural Turing Machines
Memory networks
Further reading
Summary
10. Predicting Times Sequences with Advanced RNN Chevron down icon Chevron up icon
10. Predicting Times Sequences with Advanced RNN
Dropout for RNN
Deep approaches for RNN
Stacked recurrent networks
Deep transition recurrent network
Highway networks design principle
Recurrent Highway Networks
Further reading
Summary
11. Learning from the Environment with Reinforcement Chevron down icon Chevron up icon
11. Learning from the Environment with Reinforcement
Reinforcement learning tasks
Simulation environments
Q-learning
Deep Q-network
Training stability
Policy gradients with REINFORCE algorithms
Related articles
Summary
12. Learning Features with Unsupervised Generative Networks Chevron down icon Chevron up icon
12. Learning Features with Unsupervised Generative Networks
Generative models
Semi-supervised learning
Further reading
Summary
13. Extending Deep Learning with Theano Chevron down icon Chevron up icon
13. Extending Deep Learning with Theano
Theano Op in Python for CPU
Theano Op in Python for the GPU
Theano Op in C for CPU
Theano Op in C for GPU
Coalesced transpose via shared memory, NVIDIA parallel for all
The future of artificial intelligence
Further reading
Summary
Index Chevron down icon Chevron up icon
Index

Recommendations for you

Left arrow icon
Python Machine Learning By Example
Python Machine Learning By Example
Read more
Jul 2024 518 pages
Full star icon 4.9 (8)
eBook
eBook
  • eBook eBook $9.99
  • Paperback Paperback $45.99
$9.99 $36.99
$45.99
Unlocking Data with Generative AI and RAG
Unlocking Data with Generative AI and RAG
Read more
Sep 2024 346 pages
eBook
eBook
  • eBook eBook $9.99
  • Paperback Paperback $39.99
$9.99 $31.99
$39.99
Building LLM Powered Applications
Building LLM Powered Applications
Read more
May 2024 342 pages
Full star icon 4.2 (22)
eBook
eBook
  • eBook eBook $9.99
  • Paperback Paperback $49.99
$9.99 $39.99
$49.99
Python for Algorithmic Trading Cookbook
Python for Algorithmic Trading Cookbook
Read more
Aug 2024 404 pages
Full star icon 4.6 (15)
eBook
eBook
  • eBook eBook $9.99
  • Paperback Paperback $59.99
$9.99 $47.99
$59.99
Machine Learning with PyTorch and Scikit-Learn
Machine Learning with PyTorch and Scikit-Learn
Read more
Feb 2022 774 pages
Full star icon 4.4 (95)
eBook
eBook
  • eBook eBook $9.99
  • Paperback Paperback $54.99
$9.99 $43.99
$54.99
Artificial Intelligence Engines
Artificial Intelligence Engines
Read more
Nov 2024 217 pages
eBook
eBook
  • eBook eBook $6.98
$6.98 $9.99
AI Product Manager's Handbook
AI Product Manager's Handbook
Read more
Nov 2024 484 pages
eBook
eBook
  • eBook eBook $9.99
  • Paperback Paperback $49.99
$9.99 $39.99
$49.99
RAG-Driven Generative AI
RAG-Driven Generative AI
Read more
Sep 2024 334 pages
Full star icon 4.5 (13)
eBook
eBook
  • eBook eBook $9.99
  • Paperback Paperback $43.99
$9.99 $35.99
$43.99
Hands-On Reinforcement Learning with Python
Hands-On Reinforcement Learning with Python
Read more
Jun 2018 318 pages
Full star icon 2.6 (18)
eBook
eBook
  • eBook eBook $9.99
  • Paperback Paperback $38.99
$9.99 $29.99
$38.99
LLM Engineer's Handbook
LLM Engineer's Handbook
Read more
Oct 2024 522 pages
Full star icon 4.8 (16)
eBook
eBook
  • eBook eBook $9.99
  • Paperback Paperback $59.99
$9.99 $47.99
$59.99
Right arrow icon

Customer reviews

Rating distribution
Full star icon Full star icon Full star icon Half star icon Empty star icon 3.7
(3 Ratings)
5 star 66.7%
4 star 0%
3 star 0%
2 star 0%
1 star 33.3%
Coquard Aurélien Oct 10, 2017
Full star icon Full star icon Full star icon Full star icon Full star icon 5
Don't know where to start with Neural Networks and Deep Learning? Well, here is a perfect starting point, and more ... The walkthrough all Theano concepts is illustrated with really relevant examples and algorithms. Great book! Every student in the domain should have it.
Amazon Verified review Amazon
David Oct 08, 2017
Full star icon Full star icon Full star icon Full star icon Full star icon 5
The dean from my data science graduate program told me about this book, as I am huge fan of PackT publishing. This book provides excellent resources for implementing deep learning algorithms using the Theano library in python. Plus, at the end of each chapter the author has taken the extra step of listing resent articles and publications that are relevant to that particular area of deep learning in each chapter. This one has everything that I needed to prepare to execute a deep learning computer program. It is a great resource, and so far outside of just the technical guide for Theano, there is no other book like it.
Amazon Verified review Amazon
Maciek Aug 03, 2018
Full star icon Empty star icon Empty star icon Empty star icon Empty star icon 1
This book is horrible even by Packt standards. It's full of errors, it's poorly written to the point of being barely comprehensible in some places. It almost feels like this text has been generated by a neural network rather than being written by a human author.
Amazon Verified review Amazon

People who bought this also bought

Left arrow icon
Machine Learning Engineering with Python
Machine Learning Engineering with Python
Read more
Aug 2023 462 pages
Full star icon 4.6 (36)
eBook
eBook
  • eBook eBook $9.99
  • Paperback Paperback $49.99
$9.99 $39.99
$49.99
Deep Learning with TensorFlow and Keras – 3rd edition
Deep Learning with TensorFlow and Keras – 3rd edition
Read more
Oct 2022 698 pages
Full star icon 4.6 (45)
eBook
eBook
  • eBook eBook $9.99
  • Paperback Paperback $49.99
$9.99 $39.99
$49.99
Modern Generative AI with ChatGPT and OpenAI Models
Modern Generative AI with ChatGPT and OpenAI Models
Read more
May 2023 286 pages
Full star icon 4.2 (34)
eBook
eBook
  • eBook eBook $9.99
  • Paperback Paperback $49.99
$9.99 $39.99
$49.99
Generative AI with LangChain
Generative AI with LangChain
Read more
Dec 2023 368 pages
Full star icon 4 (34)
eBook
eBook
  • eBook eBook $9.99
  • Paperback Paperback $49.99
$9.99 $39.99
$49.99
Causal Inference and Discovery in Python
Causal Inference and Discovery in Python
Read more
May 2023 456 pages
Full star icon 4.5 (49)
eBook
eBook
  • eBook eBook $9.99
  • Paperback Paperback $39.99
$9.99 $31.99
$39.99
Right arrow icon

About the author

Profile icon Bourez
Bourez
Christopher Bourez graduated from Ecole Polytechnique and Ecole Normale Suprieure de Cachan in Paris in 2005 with a Master of Science in Math, Machine Learning and Computer Vision (MVA). For 7 years, he led a company in computer vision that launched Pixee, a visual recognition application for iPhone in 2007, with the major movie theater brand, the city of Paris and the major ticket broker: with a snap of a picture, the user could get information about events, products, and access to purchase. While working on missions in computer vision with Caffe, TensorFlow or Torch, he helped other developers succeed by writing on a blog on computer science. One of his blog posts, a tutorial on the Caffe deep learning technology, has become the most successful tutorial on the web after the official Caffe website. On the initiative of Packt Publishing, the same recipes that made the success of his Caffe tutorial have been ported to write this book on Theano technology. In the meantime, a wide range of problems for Deep Learning are studied to gain more practice with Theano and its application.
Read more
See other products by Bourez
Get free access to Packt library with over 7500+ books and video courses for 7 days!
Start Free Trial

FAQs

What is the delivery time and cost of print book? Chevron down icon Chevron up icon

Shipping Details

USA:

'

Economy: Delivery to most addresses in the US within 10-15 business days

Premium: Trackable Delivery to most addresses in the US within 3-8 business days

UK:

Economy: Delivery to most addresses in the U.K. within 7-9 business days.
Shipments are not trackable

Premium: Trackable delivery to most addresses in the U.K. within 3-4 business days!
Add one extra business day for deliveries to Northern Ireland and Scottish Highlands and islands

EU:

Premium: Trackable delivery to most EU destinations within 4-9 business days.

Australia:

Economy: Can deliver to P. O. Boxes and private residences.
Trackable service with delivery to addresses in Australia only.
Delivery time ranges from 7-9 business days for VIC and 8-10 business days for Interstate metro
Delivery time is up to 15 business days for remote areas of WA, NT & QLD.

Premium: Delivery to addresses in Australia only
Trackable delivery to most P. O. Boxes and private residences in Australia within 4-5 days based on the distance to a destination following dispatch.

India:

Premium: Delivery to most Indian addresses within 5-6 business days

Rest of the World:

Premium: Countries in the American continent: Trackable delivery to most countries within 4-7 business days

Asia:

Premium: Delivery to most Asian addresses within 5-9 business days

Disclaimer:
All orders received before 5 PM U.K time would start printing from the next business day. So the estimated delivery times start from the next day as well. Orders received after 5 PM U.K time (in our internal systems) on a business day or anytime on the weekend will begin printing the second to next business day. For example, an order placed at 11 AM today will begin printing tomorrow, whereas an order placed at 9 PM tonight will begin printing the day after tomorrow.


Unfortunately, due to several restrictions, we are unable to ship to the following countries:

  1. Afghanistan
  2. American Samoa
  3. Belarus
  4. Brunei Darussalam
  5. Central African Republic
  6. The Democratic Republic of Congo
  7. Eritrea
  8. Guinea-bissau
  9. Iran
  10. Lebanon
  11. Libiya Arab Jamahriya
  12. Somalia
  13. Sudan
  14. Russian Federation
  15. Syrian Arab Republic
  16. Ukraine
  17. Venezuela
What is custom duty/charge? Chevron down icon Chevron up icon

Customs duty are charges levied on goods when they cross international borders. It is a tax that is imposed on imported goods. These duties are charged by special authorities and bodies created by local governments and are meant to protect local industries, economies, and businesses.

Do I have to pay customs charges for the print book order? Chevron down icon Chevron up icon

The orders shipped to the countries that are listed under EU27 will not bear custom charges. They are paid by Packt as part of the order.

List of EU27 countries: www.gov.uk/eu-eea:

A custom duty or localized taxes may be applicable on the shipment and would be charged by the recipient country outside of the EU27 which should be paid by the customer and these duties are not included in the shipping charges been charged on the order.

How do I know my custom duty charges? Chevron down icon Chevron up icon

The amount of duty payable varies greatly depending on the imported goods, the country of origin and several other factors like the total invoice amount or dimensions like weight, and other such criteria applicable in your country.

For example:

  • If you live in Mexico, and the declared value of your ordered items is over $ 50, for you to receive a package, you will have to pay additional import tax of 19% which will be $ 9.50 to the courier service.
  • Whereas if you live in Turkey, and the declared value of your ordered items is over € 22, for you to receive a package, you will have to pay additional import tax of 18% which will be € 3.96 to the courier service.
How can I cancel my order? Chevron down icon Chevron up icon

Cancellation Policy for Published Printed Books:

You can cancel any order within 1 hour of placing the order. Simply contact customercare@packt.com with your order details or payment transaction id. If your order has already started the shipment process, we will do our best to stop it. However, if it is already on the way to you then when you receive it, you can contact us at customercare@packt.com using the returns and refund process.

Please understand that Packt Publishing cannot provide refunds or cancel any order except for the cases described in our Return Policy (i.e. Packt Publishing agrees to replace your printed book because it arrives damaged or material defect in book), Packt Publishing will not accept returns.

What is your returns and refunds policy? Chevron down icon Chevron up icon

Return Policy:

We want you to be happy with your purchase from Packtpub.com. We will not hassle you with returning print books to us. If the print book you receive from us is incorrect, damaged, doesn't work or is unacceptably late, please contact Customer Relations Team on customercare@packt.com with the order number and issue details as explained below:

  1. If you ordered (eBook, Video or Print Book) incorrectly or accidentally, please contact Customer Relations Team on customercare@packt.com within one hour of placing the order and we will replace/refund you the item cost.
  2. Sadly, if your eBook or Video file is faulty or a fault occurs during the eBook or Video being made available to you, i.e. during download then you should contact Customer Relations Team within 14 days of purchase on customercare@packt.com who will be able to resolve this issue for you.
  3. You will have a choice of replacement or refund of the problem items.(damaged, defective or incorrect)
  4. Once Customer Care Team confirms that you will be refunded, you should receive the refund within 10 to 12 working days.
  5. If you are only requesting a refund of one book from a multiple order, then we will refund you the appropriate single item.
  6. Where the items were shipped under a free shipping offer, there will be no shipping costs to refund.

On the off chance your printed book arrives damaged, with book material defect, contact our Customer Relation Team on customercare@packt.com within 14 days of receipt of the book with appropriate evidence of damage and we will work with you to secure a replacement copy, if necessary. Please note that each printed book you order from us is individually made by Packt's professional book-printing partner which is on a print-on-demand basis.

What tax is charged? Chevron down icon Chevron up icon

Currently, no tax is charged on the purchase of any print book (subject to change based on the laws and regulations). A localized VAT fee is charged only to our European and UK customers on eBooks, Video and subscriptions that they buy. GST is charged to Indian customers for eBooks and video purchases.

What payment methods can I use? Chevron down icon Chevron up icon

You can pay with the following card types:

  1. Visa Debit
  2. Visa Credit
  3. MasterCard
  4. PayPal
What is the delivery time and cost of print books? Chevron down icon Chevron up icon

Shipping Details

USA:

'

Economy: Delivery to most addresses in the US within 10-15 business days

Premium: Trackable Delivery to most addresses in the US within 3-8 business days

UK:

Economy: Delivery to most addresses in the U.K. within 7-9 business days.
Shipments are not trackable

Premium: Trackable delivery to most addresses in the U.K. within 3-4 business days!
Add one extra business day for deliveries to Northern Ireland and Scottish Highlands and islands

EU:

Premium: Trackable delivery to most EU destinations within 4-9 business days.

Australia:

Economy: Can deliver to P. O. Boxes and private residences.
Trackable service with delivery to addresses in Australia only.
Delivery time ranges from 7-9 business days for VIC and 8-10 business days for Interstate metro
Delivery time is up to 15 business days for remote areas of WA, NT & QLD.

Premium: Delivery to addresses in Australia only
Trackable delivery to most P. O. Boxes and private residences in Australia within 4-5 days based on the distance to a destination following dispatch.

India:

Premium: Delivery to most Indian addresses within 5-6 business days

Rest of the World:

Premium: Countries in the American continent: Trackable delivery to most countries within 4-7 business days

Asia:

Premium: Delivery to most Asian addresses within 5-9 business days

Disclaimer:
All orders received before 5 PM U.K time would start printing from the next business day. So the estimated delivery times start from the next day as well. Orders received after 5 PM U.K time (in our internal systems) on a business day or anytime on the weekend will begin printing the second to next business day. For example, an order placed at 11 AM today will begin printing tomorrow, whereas an order placed at 9 PM tonight will begin printing the day after tomorrow.


Unfortunately, due to several restrictions, we are unable to ship to the following countries:

  1. Afghanistan
  2. American Samoa
  3. Belarus
  4. Brunei Darussalam
  5. Central African Republic
  6. The Democratic Republic of Congo
  7. Eritrea
  8. Guinea-bissau
  9. Iran
  10. Lebanon
  11. Libiya Arab Jamahriya
  12. Somalia
  13. Sudan
  14. Russian Federation
  15. Syrian Arab Republic
  16. Ukraine
  17. Venezuela