Skip to content

AFathi/turicreate-notebooks

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

64 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Turi Create Notebooks

This repository's main goal is to explain what Turi Create is, how to install it, and how to use it. Additionally, I'm providing a few Jupyter Notebook examples that demonstrate different uses of the Python package.

Also, in this README file, I will explain how to setup Python, pip, and install the turicreate package on your machine.

Table of Contents Description
What is Turi Create? Describes what Turi Create is and what developers can use it for.
Why use Turi Create? Describes why Turi Create is useful for developers.
Requirements Describes the machine and software requirements needed to use Turi Create and run the Jupyter Notebook examples.
Install & Setup Explains the installation and setup process in details.
Contributions Describes how you can contribute to this project.
Turi Create Docs Redirects you to the official Turi Create documentations.
License Describes the license of this project.

What is Turi Create?

In the Turi Create official repository, they describe it as a library that:

simplifies the development of custom machine learning models. You don't have to be a machine learning expert to add recommendations, object detection, image classification, image similarity or activity classification to your app.

From a point of view of an iOS developer with some background in machine learning, I'd describe it as a library that does the complex mathematical computation for you, while you only worry about finding the right dataset for your machine learning model.

Some machine learning models created using Turi Create can easily be exported as a Core ML model and be implemented into iOS, macOS, tvOS or watchOS applications.

Models that easily allow Core ML exporting include:

  • Image Classification models
  • Object Detection models

Why use Turi Create?

In the past few years, the use of machine learning approaches to solve problems and perform complex tasks have been increasing. Machine learning enable us to use big data to perform complex tasks, such as image classification.

Before Turi Create, it would be very challenging and time consuming for software developers to create a machine learning model and run it on mobile devices. In order to run a machine learning model on a mobile device, you would have to optimize your model and make it work on devices that doesn't have the hardware capabilities, such as the iPhone.

However, thanks to GraphLab and Apple, now we have both Turi Create and CoreML that enable us to easily create machine learning models for Apple devices. Turi Create provides us with essential machine learning algorithms, such as k-nearest neighbor, and advanced deep learning algorithms, such as Residual Networks (ResNet), in order to create a production-ready machine learning trained model.

In this repository, I will be demonstrating how to create useful machine learning models that may have many uses in your existing and future applications!

Requirements

In order to install, setup and try out the example notebooks you will need the following:

  • A computer with a 64-bit processor (x86_64 architecture)
  • Anaconda, Python 2.7 version: Download here

Note:- Currently, Turi Create only supports Python 2.7.

Install & Setup

After making sure our machine has x86_64 architecture we will do the following:

  1. Download & install Anaconda, Python 2.7 version
  2. After done installing Anaconda, go to your computer's terminal and create an environment to install Turi Create, by doing the following:
$ conda create -n turienv
  1. Activate your new enviornment:
source activate turienv
  1. Install Turi Create:
pip install -U turicreate
  1. Restart Anaconda-Navigator then launch Jupyter Notebook

  2. Clone this repository and redirect to it through Jupyter Notebook!

Check How to run Python Notebooks? and How to run Python? for more details on testing the examples on your machine.

Contributions

If you have an idea for a new Turi Create example and want to add it to this repository, feel free to fork the project and create a pull request!

Also, feel free to create an issue if you have any suggestions or need any help ☺️

License

This repository is published under the MIT license. Check LICENSE for more information.

Note:- Turi Create is under a BSD-3-Clause license. Check the official repository LICENSE for more information.

Releases

No releases published

Packages

No packages published