Repositories

https://github.com/apple/turicreate

Introduction

What is Turi Create?

Turi Create is a tool used to simplify the development of custom machine learning models. Turi Create helps in developing custom machine learning whereas you don’t need to be an expert of machine learning before you could be able to add recommendations, object detecting, classifying of image and classification of activity to your application.

Components

My suggestion is on the

• Custom Machine Learning

PROPOSAL

Using Turi Create can help to accomplish any common ML task

Proposing the addiction/BENEFIT :

• RECOMMENDER:

Turi Create develop the option of personalizing choices for users where they have privilege of choosing the activity of their choice.

• CLASSIFICATION OF IMAGE

Turi Create is specialize in labeling various image for users to be able differentiate image in activity.

• STYLE TRANSFER

Turi Create beautify image in application by stylizing different image to suit the users interest.

• IMAGE SIMILARITY

Turi Create perform the task of finding similar images.

• CLUSTERING

Turi Create perform the task of grouping similar data point and merge them together.

• ACTIVITY CLASSIFICATION

Turi Create make use of sensors in order to detect activity on application during its use.

• REGRESSION

Turi Create also predict numeric values in an application.

• TEXT CLASSIFIER

Turi Create perform the functions of analyzing sentiment of different messages

This will be explained further in the mockups below.

MOCKUP EXAMPLES

Users that intends to make his/her app to recognize a specific objects on image. Model could be built with just a few line of code.

import turicreate as tc

Load data

data = tc.SFrame('photoLabel.sframe')

Create a model

model = tc.image_classifier.create(data, target='photoLabel')

Make predictions

predictions = model.predict(data)

Export to Core ML

model.export_coreml('MyClassifier.mlmodel')

The above statement personalize is an example of a built line of code in order for the app to recognize a specific object on image

Turi Create FEATURES

Easy to use-

It is a form of simplicity in the sense that it make it easy for users because it just focus on the main task instead of its algorithms.

Flexible-

It is flexible in the sense that it support both the text, images, audio, video and its sensor data

Fast and Scalable –

The work rate is at a high speed and standard and it works with a large datasets perfect with just a single machine.

Visual-

Streaming of visual is at a high level of standardization in order for a perfection of exploring data.

Ready to Deploy –

Models is being deploy successfully to Core ML for use in IOS, MacOS, watchOS ad tvOS apps.

Conclusion

The application is undergoing a constant development and using this application will improve many user machine learning uses and would be easy for all instead of being an expert in machine learning.
It will be free and easy to use for every users without the need of any difficulty steps.

Supported Platforms

Turi Create supports:
macOS 10.12+
Linux (with glibc 2.12+)
Windows 10 (via WSL)
System Requirements

Turi Create requires:

Python 2.7, 3.5, 3.6
x86_64 architecture
At least 4 GB of RAM

My GitHub profile

https://github.com/adewararilwan