/SG_AIMLPros-MOST

This repository consist of SG_AIMLPros project for Project showcase challenge in Udacity Facebook SPAIC 2019

SPAIC | SG_AIMLPros

SPAIC - Udacity | Facebook Secure and Privacy AI Challenge 2019.

Project MOST (SG_AIMLPros-MOST/Projects)

Purpose 1: MOST - MOdeling and STatistics

Implementing different models on same dataset | Providing statistics on "which is the most accurate model for image processing" through the details collected from the models implemented

Purpose 2:

Rice is the staple food for more than half the world's population;accordingly, its supply must double by 2050 to keep up with food demand from population growth. Nearly a fifth of the world’s population are rice farmers.However,they approximately loose 37% of their rice crop yields to various rice diseases. We believe our model will help rice farmers know their rice crop’s condition ,hence leading to higher yields. For more detailed project introduction: https://github.com/risper25/MOST- (@risper bevalyn)

Project Participants (Total count : 7)

This is the slack handles of the members

Name Slack Name
Vigneshwari Ramakrishnan @Vigneshwari
Shudipto Trafder @Shudipto Trafder
Laura Truncellito @LauraT
Ellyana Linden @Ellyana Linden
Risper Bevalyn Adera @risper bevalyn
Shuvam Lal @happycoder354
Anna Scott @Anna Scott

Contributions made by the team

1) Pre-trained Model implementation

  1. @Ellyana Linden
  2. @Shudipto Trafder
  3. @Vigneshwari
  4. @risper bevalyn
  5. @happycoder354

2) Technical Documentations | Other

Slack Name Contribution Areas
@LauraT Description of the models implemented for the projects, Readme file
@Vigneshwari Readme file, Project deck, Github Project repo, Portfolio
@risper bevalyn Introduction of the project, Data collection and chart in google sheet
@Shudipto Trafder Essential attributes identification and listing, Readme file

Dataset Utilized

Datasets: Rice Diseases Image Dataset

Details of the dataset

Dataset of rice images - 4 labels/ categories (3 diseases and pests, 1 healthy).

  1. Brown Spot (523 files)
  2. Healthy (1488 files)
  3. Hispa (565 files)
  4. LeafBlast (779 files)

Pretrained Models used in this project

  1. GoogleNet
  2. ResNet50
  3. VGG19
  4. Resnext101_32x8d
  5. Densenet201
  6. Alexnet

Link of our project deck / activity records

Link: https://github.com/drvigneshwari/SG_AIMLPros-MOST/projects/1

Link of all models implemented by the participants

Slack Name Model Name Kernel Link
@Ellyana Linden GoogleNet googlenet.ipynb
@Ellyana Linden Resnet50 googlenet
@Shudipto Trafder resnext101_32x8d kernelcd66c7e01d-version1
@Shudipto Trafder resnet 50 kernelcd66c7e01d
@Vigneshwari Alexnet kernel4946364e36
@risper bevalyn, @happycoder354 VGG19 rice-leaves-disease-classifier

Resources used for reference