Primary competition visual

ICLR Workshop Challenge #1: CGIAR Computer Vision for Crop Disease

Helping East Africa
$5 000 USD
Completed (almost 6 years ago)
Classification
Computer Vision
1050 joined
338 active
Starti
Jan 29, 20
Closei
Mar 28, 20
Reveali
Mar 29, 20
Starter Code - Using Monk transfer learning low-code library - score: 0.45 (approx)
Notebooks · 2 Mar 2020, 17:35 · 8

Added a starter code example using Monk library: https://github.com/Tessellate-Imaging/monk_v1/blob/master/tutorials/zindi_starter_code_transfer_learning_with_monk_score_0.45.ipynb

Library: https://github.com/Tessellate-Imaging/monk_v1

Details: Used Densenet201, batch_size of 8 images, trained for 20 epochs on 980 TI GPU

Info - 1: The library has a project management feature which allows you to compare different experiments carried out by changing hyper-parameters and base models. See an example here - https://medium.com/@monkai.betasignup/fine-tuning-with-monk-6cec33fd6252

Info - 2: The libray has a feature to easily find out right hyper-parameters. See and example here - https://medium.com/@monkai.betasignup/plant-disease-classification-with-monk-3a4d7ba419b9

It's an opensource library, so do star it if you find it useful for your project .

Happy Coding!!!

Discussion 8 answers

Thank you for sharing this @abhishek4273

2 Mar 2020, 19:11
Upvotes 0

@abhishek4273 It looks like the link you provided is broken. Is the code still available anywhere by any chance? Thanks!

15 Mar 2020, 07:33
Upvotes 0

Hi, we restructured our library so the link changed. Here's the updated link: https://github.com/Tessellate-Imaging/monk_v1/blob/master/study_roadmaps/4_image_classification_zoo/Zindi%20Competition%20-%20Plant%20Disease%20Classification.ipynb