rafka-imandaputra / orbit-tsdn-2022 Goto Github PK

Built by PyTorch, Flask, Albumentations, WandB, ResNet34 with support by JavaSript and Tableau Public.

The ORBIT team from SMKN 4 Bandung wanted to innovate in agriculture use cases especially paddy production. We created this project to take part in Turnamen Sains Data Nasional, one of the prestigious annual data science competitions in Indonesia which is supported by many government agencies and universities in Indonesia. This year's TSDN theme is Data Science and AI Implementation for Business.

Our goal is to seek insight from an analytical dashboard that provides solutions to be followed up by creating computer vision models whose results are implemented in web apps. So we generate 3 outputs with the following details:

Our project is divided into 3 related parts

1. Dashboard Analytics
   Generate insights that become solutions to be followed up.
   visit https://public.tableau.com/app/profile/raka.iskandar/viz/VisualisasiDataTSDN2022/Storyboard to see dashboard with interactive view.
2. Computer Vision
   Following up on the findings of the dashboard into a computer vision model.
   visit https://www.kaggle.com/code/rafkaip/orbit-cv/ to run or copy notebook.
3. Web Apps
   Implementing the computer vision model that has been made.
   visit https://orbit-padi-kita.herokuapp.com/ to browse the deployed website or visit https://github.com/raihanaiooo/PadiKita/tree/master to see the full source code.

All credits belong to the data source:

1. Food and Agriculture Organization of United Nations. (2022). FAOSTAT. https://www.fao.org/faostat/en/#data/QCL
   
2. Department of Computer Science and Engineering, Manonmaniam Sundaranar University, Tirunelveli, India. (2022). Paddy Doctor: Paddy Disease Classification. https://www.kaggle.com/competitions/paddy-disease-classification/data.