TuriCreate(Machine Learning Python Library):
Custom Machine Learning models can be made using Turi create very easily. You don’t have to be an expert in Machine Learning to use Turi create and it helps in building an add recommendations, object detection, image classification, image similarity or activity classification systems easily.
Advantages of TuriCreate:
- It is easy to use.
- Flexible to use.
- Ready to deploy model to ios, MacOs, watchOs and tvOs apps.
- It one of the crazy features is that it can read the .csv(Comma Separated Values) and use the Machine Learning Model.
Supported Platform for TuriCreate:
- macOS 10.12+
- Linux (with glibc 2.12+)
- Windows 10 (via WSL)=It is little tricky to install it on windows but you can install via WSL(Windows Subsystem Linux) and if does have the same error then you have to upgrade pip by using this command:
- pip3 install –upgrade pip
- System Requirements
- Python 2.7, 3.5 or above (ex. 3.7, 3.8)
- Python 3.7 macOS only+
- x86_64 architecture
Using a Conda Environment or virtualenv is highly recommended so that we can have different Python Interpreter on the same Machine.
Turicreate uses SFrames for data handling:
SFrames means scalable data frame. It is a tabular, column-mutable data frame object that can scale to big data. It’s one of the advantages is that is mutable.
Code: How to use SFrame for Loading the data:
import turicreate as tc from turicreate import SFrame # Let's declare a variable for loading the data data = tc.SFrame("data.csv") # Here data.csv is a file stored in your system # with name = data |
Data Structure of TuriCreate:
- SFrame
- SGraph
- SArray
