https://github.com/n-y-kim/predicttestscoreapp
iOS Application for the 'Prediction of test score' with scikit & coreML
https://github.com/n-y-kim/predicttestscoreapp
coreml data-science googlecolab ios jupyter-notebook machine-learning prediction sklearn xcode
Last synced: 9 months ago
JSON representation
iOS Application for the 'Prediction of test score' with scikit & coreML
- Host: GitHub
- URL: https://github.com/n-y-kim/predicttestscoreapp
- Owner: n-y-kim
- License: mit
- Created: 2021-08-26T12:26:34.000Z (over 4 years ago)
- Default Branch: main
- Last Pushed: 2021-08-28T09:28:08.000Z (over 4 years ago)
- Last Synced: 2025-05-12T08:09:42.501Z (11 months ago)
- Topics: coreml, data-science, googlecolab, ios, jupyter-notebook, machine-learning, prediction, sklearn, xcode
- Language: Jupyter Notebook
- Homepage:
- Size: 4.49 MB
- Stars: 0
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Data-Science & Machine Learning(scikit, coreML) to iOS Application
Project inspired from ['Dongguk Uni X Naver Boostcourse'](https://www.boostcourse.org/study-dongguk-1)
Prediction of test score with [dataset](https://www.kaggle.com/kwadwoofosu/predict-test-scores-of-students)
### Application
### Data Analysis
Working on a application that can predict student's score tests(posttest) by
* School
* School Setting: Urban / Rural / Suburban
* School Type: Public / Non-public
* Classroom
* Teaching Method : Standard / Experimental
* Number of Students
* Gender: Male / Female
* Lunch: Qualified / Not qualified
## This repository contains
* .csv file of the dataset file
* .keynote of the project(mostly about the data analysis)
* .ipynb file from Google Colab & Jupyter Notebook
* .xcodeproj file of the application
## Beta Version Explanation
👌🏻 Okay and DONE:
* Data Analysis of the dataset
* used Linear Regression(scikit) to predict posttest scores without pretest in train data. ➡️ 95% score
* Prototype of the application(UI)
* conversion of model to CoreML(.mlmodel)
❌ NOT OKAY and need UPDATES:
* The .ipynb files need to be organized(it may be hard to figure out the purpose of the code. :( This is my bad!!)
* The model(scores.mlmodel) show wrong output if the combination of input datas are not from trained data => how should I fix?
* The model inputs are extremely difficult to get their data as the DataFrame went through One-Hot-Encoding
## Track Update
beta 0.1 version
## Feedbacks & Contacts
📮 nyn2265@gmail.com
⭐️ issue or discussion