CoreML - Master Machine Learning for iOS Apps

Name: CoreML - Master Machine Learning for iOS Apps
Rating: 4.5 (158 reviews)

Build Intelligent iOS 11 apps using Apple's Native Machine Learning API - CoreML

Created byApp Brewery Co.

Last updated 2/2018

English

English [Auto],

What you'll learn

Build intelligent iOS apps that use machine learning.
Create an app that uses the Inception V3 framework with CoreML to identify objects visually.
Understand how the ImagePickerView works and how you can enable the camera functionality in your iOS apps.
Make an app that recognises the name of a flower just from a picture.

Course content

4 sections • 36 lectures • 3h 48m total length

Install the Required Software0:15
Introduction to Machine Learning5:02
Learn what machine learning is and how Apple's new framework enables apps to learn from data. Apply visual recognition in an iOS app using Swift.
Supervised Learning14:09
Discover how supervised learning uses labeled training data to teach models to classify new inputs, and how regression and classification apply to examples like spam filters and salaries.
Unsupervised Learning6:09
Explore unsupervised learning, where models find structure in unlabeled data through clustering, even in social datasets like Facebook. They learn to identify data groups and patterns without explicit labels.
Reinforcement Learning7:14
Compare reinforcement learning with supervised and unsupervised learning. See how AlphaGo, trained on thousands of games, uses reinforcement learning and deep learning to optimize moves toward a winning outcome.

What is CoreML?6:51
Core ML loads pre trained models and runs on-device predictions for classification or regression, using dot eml model files and ready-to-use models like Inception V3.
What We'll Make By the End of This Module2:21
Build a hot dog or not hot dog image classifier for an iOS app using the camera to capture photos and reveal the classification.
How to Import the Image Recognition Models and Enable Camera Functionality21:55
Learn to import a pre trained image recognition model (Inception V3) into an iOS project with CoreML and Vision, and enable camera-based image capture via a UI image picker.
Getting Image Recognition Results Back from Our Inception V3 Model14:24
Convert user-selected UI image to a CIImage, run it through the Inception V3 Core ML model via the Vision framework, and interpret the classification results.
Hotdog or Not Hotdog?7:16
Classify a user-selected image with the Inception V3 model using Vision and CoreML, then set the navigation title to hot dog or not hot dog based on the top result.
Download the Completed App Project0:08

What We'll Make By the End of This Module4:01
Convert a pre-trained model into a dot eml model file using Python tools and load it into your Xcode project to create an offline flower recognition app.
Installing CoreML Tools Using Python9:18
Master the setup of pip, create a python two point seven virtual environment, and install Corrimal Tools with pip to enable CoreML workflows for iOS apps.
Converting a Caffe Model in to MLModel11:58
Convert a caffe model into a CoreML mlmodel via a Python script, load the model files and labels, and save the flower classifier as a CoreML model.
[Challenge] Set up a UIImagePickerController1:41
Import the flower classifier mlmodel into a new Xcode project, then configure a UIImagePickerController to capture a photo and display it in an image view using a camera button.
[Solution] Set up a UIImagePickerController5:54
Set up a UIImagePickerController to capture photos with the camera, embed in a navigation controller, configure privacy, handle the delegate, present and dismiss the picker, and enable optional cropping.
[Challenge] Getting Classifications from our Model1:00
Use the flower classifier to classify the image from the image picker and update the navigation bar title with the detected flower name via a Detect method.
[Solution] Getting Classifications from our Model9:00
Learn to build an iOS image classification flow with Core ML and Vision, converting user images to CI image and classifying flowers with a flower classifier model.
[Challenge] Adding Cocoapods to Our Project3:05
Add Wikipedia lookups to your CoreML flower recognizer by integrating Alamofire for REST API calls and CocoaPods for dependency management, specifying versions, and installing them to fetch flower information.
[Solution] Adding Cocoapods to Our Project1:48
Configure CocoaPods in your project by creating a podfile, running pod install, and opening the generated Xcode workspace to verify Alamofire and SwiftyJSON imports.
[Challenge] Making HTTP Get requests to Wikipedia using Alamofire and REST6:14
Learn to construct a Wikipedia API call with Alamofire to fetch the Barberton daisy intro extract as JSON, using endpoint, parameters, format=json, action=query, extract and titles with redirect=true.
[Solution] Making HTTP Get requests to Wikipedia using Alamofire and REST3:42
Use Alamofire to perform a get request to Wikipedia with parameters from a text file, handle the json response, and display the Wikipedia extract in a label.
[Challenge] Parsing the JSON Result using SwiftyJSON2:09
Take a data response from a Wikipedia fetch, convert it to JSON with the Swift JSON framework, then parse it with SwiftyJSON to display the extract in a label.
[Solution] Parsing the JSON Result using SwiftyJSON5:56
Learn to parse a Wikipedia JSON response with Alamofire and SwiftyJSON, extract the article text via query, pages, pageid, and display it in a label while refining the UI.
Displaying Images from a URL using SDWebImage5:15
Pull an image from a URL using SDWebImage, specify the page image thumbnail size to 500 by 500 from Wikipedia, and display the Barberton daisy in the app's image view.
[Bonus Lesson] Creating the App Icon from Scratch and Styling the App6:47
Style the app with a garden theme, adjust imagery and colors, and create a flower app icon from scratch using Illustrator and Photoshop, preparing CoreML assets for flower classification.
Download the Completed App Project0:08

What We'll Make By the End of This Module1:16
Learn to build a visual recognition app with Swift using IBM blue mics, capture photos with a camera, interpret results, and share findings on Twitter.
A Note About RestKit0:16
How to Install and Use Carthage as a Dependency Manager10:07
Install Carthage as a dependency manager and integrate the IBM Watson visual recognition Swift SDK into an iOS Xcode project, including obtaining the API key.
The Many Ways of Installing Carthage0:24
How to Implement the Camera and Photo Library Functionality14:19
Integrate a ui image picker to access the camera or photo library, then display the selected image in a full-screen image view and handle permissions.
How to Use a Pre-Trained Neural Net from IBM to Classify Any Image8:51
Tap into a pre-trained IBM visual recognition model to classify images in an iOS app, wiring the UI, API key, and version, and test results in the simulator.
How to Process the Data You Get Back from IBM Bluemix10:04
Learn to process IBM Watson classification results, extract the top items, check for hot dog, and update the iOS app UI on the main thread.
Let's Beautiful Our App and Make it Look like the Real Deal19:02
Polish the app by disabling the camera button during classification and showing a spinner, then enable sharing once hot dog or not hot dog results appear.
Download the Completed App Project0:08

Requirements

Familiar with basic iOS app development
A Mac or PC running macOS Sierra 10.12.4 or later
No paid software required - all apps will be created in Xcode 9 (which is free)

Description

Have an app idea that requires machine learning?

But all the maths getting you down?

Want a simple and easy way of implementing machine learning in your iOS apps?

Then this course is for you!

Join me as we dive into Apple's latest iOS 11 API - CoreML - a native iOS framework built with Swift.

After this course you fully understand:

What is machine learning
Supervised learning
Unsupervised learning
Reinforcement learning
How CoreML works
How to implement image recognition in your iOS apps
How to convert existing trained models created using Caffe, Keras, TensorFlow or scikit Learn into a MLModel that you can drag and drop into your iOS apps.
and much much more...

Take a look at what other students have been saying about this course:

"I am about a third through this course and I have learned so much. This course is worth way more than what it cost but I'm thankful prices are low or I might have passed it up in the first place not knowing what I would get. I have used a couple Udemy courses and countless youtube tutorials. This is the best course I've ever took." - Jeffrey Nelson

"The course offers interesting concepts coupled with a teacher that explains things clearly. You get to make a bunch of interesting apps and expand your skills. Thanks Angela!" - Megan Eding

"Clear tutorials, the lecturer explains everything well. It is one of the best courses I have participated in on Udemy." - Mark Garcia

"Angela is connecting the missing dots for me... " - Tom Colenbrander

What are you waiting for? Join the course today!

Who this course is for:

If you are an intermediate to advanced iOS developer looking to add more skills under your belt.
If you are looking to quickly get up to date with the latest iOS 11 APIs.
If you are a beginner and want to learn to create intelligent iOS apps using machine learning without having to understand a lot of complicated maths.

CoreML - Master Machine Learning for iOS Apps

What you'll learn

Explore related topics

Course content

Introduction to Machine Learning5 lectures • 33min

Apple's Latest API for iOS 11 - CoreML Framework6 lectures • 53min

Advanced CoreML - How to Convert Pre-Trained Models Using CoreML Tools16 lectures • 1hr 18min

[Bonus Section] Adding Machine Learning to iOS Apps Using IBM Watson9 lectures • 1hr 4min

Requirements

Description

Who this course is for: