Machine Learning Projects for Mobile Applications

Machine Learning Projects for Mobile Applications

Bring magic to your mobile apps using TensorFlow Lite and Core ML Key Features Explore machine learning using classification, analytics, and detection tasks. Work with image, text and video datasets to delve into real-world tasks Build apps for Android and iOS using Caffe, Core ML and Tensorflow Lit...

Descripción completa

Detalles Bibliográficos
Otros Autores: NG, Karthikeyan, author (author)
Formato: Libro electrónico
Idioma:Inglés
Publicado: Packt Publishing 2018.
Edición:1st edition
Ver en Biblioteca Universitat Ramon Llull:https://discovery.url.edu/permalink/34CSUC_URL/1im36ta/alma991009631835006719
Tabla de Contenidos:
  • Cover
  • Title Page
  • Copyright and Credits
  • Dedication
  • Packt Upsell
  • Contributors
  • Table of Contents
  • Preface
  • Chapter 1: Mobile Landscapes in Machine Learning
  • Machine learning basics
  • Supervised learning
  • Unsupervised learning
  • Linear regression - supervised learning
  • TensorFlow Lite and Core ML
  • TensorFlow Lite
  • Supported platforms
  • TensorFlow Lite memory usage and performance
  • Hands-on with TensorFlow Lite
  • Converting SavedModel into TensorFlow Lite format
  • Strategies
  • TensorFlow Lite on Android
  • Downloading the APK binary
  • TensorFlow Lite on Android Studio
  • Building the TensorFlow Lite demo app from the source
  • Installing Bazel
  • Installing using Homebrew
  • Installing Android NDK and SDK
  • TensorFlow Lite on iOS
  • Prerequisites
  • Building the iOS demo app
  • Core ML
  • Core ML model conversion
  • Converting your own model into a Core ML model
  • Core ML on an iOS app
  • Summary
  • Chapter 2: CNN Based Age and Gender Identification Using Core ML
  • Age, gender, and emotion prediction
  • Age prediction
  • Gender prediction
  • Convolutional Neural Networks
  • Finding patterns
  • Finding features from an image
  • Pooling layer
  • Rectified linear units
  • Local response normalization layer
  • Dropout layer
  • Fully connected layer
  • CNNs for age and gender prediction
  • Architecture
  • Training the network
  • Initializing the dataset
  • The implementation on iOS using Core ML
  • Summary
  • Chapter 3: Applying Neural Style Transfer on Photos
  • Artistic neural style transfer
  • Background
  • VGG network
  • Layers in the VGG network
  • Building the applications
  • TensorFlow-to-Core ML conversion
  • iOS application
  • Android application
  • Setting up the model
  • Training your own model
  • Building the application
  • Setting up the camera and an image picker
  • Summary
  • References.
  • Chapter 4: Deep Diving into the ML Kit with Firebase
  • ML Kit basics
  • Basic feature set
  • Building the application
  • Adding Firebase to our application
  • Face detection
  • Face orientation tracking
  • Landmarks
  • Classification
  • Implementing face detection
  • Face detector configuration
  • Running the face detector
  • Step one: creating a FirebaseVisionImage from the input
  • Using a bitmap
  • From media.Image
  • From a ByteBuffer
  • From a ByteArray
  • From a file
  • Step two: creating an instance of FirebaseVisionFaceDetector object
  • Step three: image detection
  • Retrieving information from detected faces
  • Barcode scanner
  • Step one: creating a FirebaseVisionImage object
  • From bitmap
  • From media.Image
  • From ByteBuffer
  • From ByteArray
  • From file
  • Step two: creating a FirebaseVisionBarcodeDetector object
  • Step three: barcode detection
  • Text recognition
  • On-device text recognition
  • Detecting text on a device
  • Cloud-based text recognition
  • Configuring the detector
  • Summary
  • Chapter 5: A Snapchat-Like AR Filter on Android
  • MobileNet models
  • Building the dataset
  • Retraining of images
  • Model conversion from GraphDef to TFLite
  • Gender model
  • Emotion model
  • Comparison of MobileNet versions
  • Building the Android application
  • References
  • Questions
  • Summary
  • Chapter 6: Handwritten Digit Classifier Using Adversarial Learning
  • Generative Adversarial Networks
  • Generative versus discriminative algorithms
  • Steps in GAN
  • Understanding the MNIST database
  • Building the TensorFlow model
  • Training the neural network
  • Building the Android application
  • FreeHandView for writing
  • Digit classifier
  • Summary
  • Chapter 7: Face-Swapping with Your Friends Using OpenCV
  • Understanding face-swapping
  • Steps in face-swapping
  • Facial key point detection
  • Identifying the convex hull.
  • Delaunay triangulation and Voronoi diagrams
  • Affine warp triangles
  • Seamless cloning
  • Building the Android application
  • Building a native face-swapper library
  • Android.mk
  • Application.mk
  • Applying face-swapping logic
  • Building the application
  • Summary
  • References
  • Questions
  • Chapter 8: Classifying Food Using Transfer Learning
  • Transfer learning
  • Approaches in transfer learning
  • Training our own TensorFlow model
  • Installing TensorFlow
  • Training the images
  • Retraining with own images
  • Training steps parameter
  • Architecture
  • Distortions
  • Hyperparameters
  • Running the training script
  • Model conversion
  • Building the iOS application
  • Summary
  • Chapter 9: What's Next?
  • What you have learned so far
  • Where to start when developing an ML application
  • IBM Watson services
  • Microsoft Azure Cognitive Services
  • Amazon ML
  • Google Cloud ML
  • Building your own model
  • Limitations of building your own model
  • Personalized user experience
  • Better search results
  • Targeting the right user
  • Summary
  • Further reading
  • Other Books You May Enjoy
  • Index.

Ejemplares similares