IOS predict prices

author: steeve laquitaine

Description: This is the codebase for a small iOS app that reports price predictions based on various house variables such as the number of rooms, etc.,.. The predictions are calculated by a machine learning model trained on the Boston housing dataset.

setup

conda create ios_predict python==3.6.2
conda activate ios_predict

model building

python src/build_model.py 
# produces bhousing.mlmodel in project's root

Create xcode project and add model

1. open xcode
2. select single view app - ios - swiftUI (no need to select a team: None)
3. Manually move bhousing.mlmodel to core_ml_demo/ folder

Edit contentView.swift

contentView.swift must be programmed with Apple's Swift programming language.

Build and run

• select e.g., simulator iphone 11 Pro Max
• build (Ctrl + r) or click play

A interactive simulator phone will appear to preview and test the app