UID:
kobvindex_INTEBC5573403
Format:
1 online resource (244 pages)
Edition:
1st ed.
ISBN:
9781788623087
Content:
This book gives you a practical, hands-on understanding of how you can leverage the power of Python and Keras to perform effective deep learning. It presents a unique problem-solution approach to tackle various problems in training different types of neural networks while taking care of the speed and accuracy of these models
Note:
Cover -- Title Page -- Copyright and Credits -- Packt Upsell -- Contributors -- Table of Contents -- Preface -- Chapter 1: Keras Installation -- Introduction -- Installing Keras on Ubuntu 16.04 -- Getting ready -- How to do it... -- Installing miniconda -- Installing numpy and scipy -- Installing mkl -- Installing TensorFlow -- Installing Keras -- Using the Theano backend with Keras -- Installing Keras with Jupyter Notebook in a Docker image -- Getting ready -- How to do it... -- Installing the Docker container -- Installing the Docker container with the host volume mapped -- Installing Keras on Ubuntu 16.04 with GPU enabled -- Getting ready -- How to do it... -- Installing cuda -- Installing cudnn -- Installing NVIDIA CUDA profiler tools interface development files -- Installing the TensorFlow GPU version -- Installing Keras -- Chapter 2: Working with Keras Datasets and Models -- Introduction -- CIFAR-10 dataset -- How to do it... -- CIFAR-100 dataset -- How to do it... -- Specifying the label mode -- MNIST dataset -- How to do it... -- Load data from a CSV file -- How to do it... -- Models in Keras - getting started -- Anatomy of a model -- Types of models -- Sequential models -- How to do it... -- Create a Sequential model -- Compile the model -- Train the model -- Evaluate the model -- Predict using the model -- Putting it all together -- Model inspection internals -- Model compilation internals -- Initialize the loss -- Model training -- Output of the sample -- Shared layer models -- Introduction - shared input layer -- How to do it... -- Concatenate function -- Keras functional APIs -- How to do it... -- The output of the example -- Keras functional APIs - linking the layers -- How to do it... -- Model class -- Image classification using Keras functional APIs -- How to do it
,
Chapter 3: Data Preprocessing, Optimization, and Visualization -- Feature standardization of image data -- Getting ready -- How to do it... -- Initializing ImageDataGenerator -- Sequence padding -- Getting ready -- How to do it... -- Pre-padding with default 0.0 padding -- Post-padding -- Padding with truncation -- Padding with a non-default value -- Model visualization -- Getting ready -- How to do it... -- Code listing -- Optimization -- Common code for samples -- Optimization with stochastic gradient descent -- Getting ready -- How to do it... -- Optimization with Adam -- Getting ready -- How to do it... -- Optimization with AdaDelta -- Getting ready -- How to do it... -- Adadelta optimizer -- Optimization with RMSProp -- Getting ready -- How to do it... -- Chapter 4: Classification Using Different Keras Layers -- Introduction -- Classification for breast cancer -- How to do it... -- Data processing -- Modeling -- Full code listing -- Classification for spam detection -- How to do it... -- Data processing -- Modeling -- Full code listing -- Chapter 5: Implementing Convolutional Neural Networks -- Introduction -- Cervical cancer classification -- Getting ready -- How to do it... -- Data processing -- Modeling -- Predictions -- Digit recognition -- Getting ready -- How to do it... -- Modeling -- Chapter 6: Generative Adversarial Networks -- Introduction -- GAN overview -- Basic GAN -- Getting ready -- How to do it... -- Building a generator -- Building a discriminator -- Initialize the GAN instance -- Training the GAN -- Output plots -- Average metrics of the GAN -- Boundary seeking GAN -- Getting ready -- How to do it... -- Generator -- Discriminator -- Initializing the BGAN class -- Boundary seeking loss -- Train the BGAN -- Output the plots -- Iteration 0 -- Iteration 10000 -- Metrics of the BGAN model -- Plotting the metrics -- DCGAN
,
Getting ready -- How to do it... -- Generator -- Summary of the generator -- Training the generator -- Discriminator -- Build the discriminator -- Summary of the discriminator -- Compile the discriminator -- Combined model - generator and discriminator -- Train the generator using feedback from a discriminator -- Putting it all together -- The output of the program -- Average metrics of the model -- Chapter 7: Recurrent Neural Networks -- Introduction -- The need for RNNs -- Simple RNNs for time series data -- Getting ready -- Loading the dataset -- How to do it... -- Instantiate a sequential model -- LSTM networks for time series data -- LSTM networks -- LSTM memory example -- Getting ready -- How to do it... -- Encoder -- LSTM configuration and model -- Train the model -- Full code listing -- Time series forecasting with LSTM -- Getting ready -- Load the dataset -- How to do it... -- Instantiate a sequential model -- Observation -- Sequence to sequence learning for the same length output with LSTM -- Getting ready -- How to do it... -- Training data -- Model creation -- Model fit and prediction -- Chapter 8: Natural Language Processing Using Keras Models -- Introduction -- Word embedding -- Getting ready -- How to do it... -- Without embeddings -- With embeddings -- Sentiment analysis -- Getting ready -- How to do it... -- Full code listing -- Chapter 9: Text Summarization Using Keras Models -- Introduction -- Text summarization for reviews -- How to do it... -- Data processing -- Encoder-decoder architecture -- Training -- See also -- Chapter 10: Reinforcement Learning -- Introduction -- The CartPole game with Keras -- How to do it... -- Implementing the DQN agent -- The memory and remember -- The replay function -- The act function -- Hyperparameters for the DQN -- DQN agent class -- Training the agent -- Dueling DQN to play Cartpole -- Getting ready
,
DQN agent -- init method -- Setting the last layer of the network -- Dueling policy -- Init code base -- BoltzmannQPolicy -- Adjustment during training -- Sequential memory -- How to do it... -- Plotting the training and testing results -- Other Books You May Enjoy -- Index
Additional Edition:
Print version Dua, Rajdeep Keras Deep Learning Cookbook Birmingham : Packt Publishing, Limited,c2018 ISBN 9781788621755
Language:
English
Keywords:
Electronic books
URL:
Full-text
((OIS Credentials Required))
Bookmarklink
