sugarcane_disease_pred

[IkkiOcean]

Pull Request

Title

Ensemble Model for Sugarcane Disease Prediction

Description

This pull request implements an ensemble model combining MobileNet, InceptionV3, VGG16, Conv2D, and ResNet152 for predicting sugarcane diseases. The model achieves an accuracy of 90.94%, significantly improving upon the individual models.

Related Issues

Closes : #109

Changes Made

• Added a Jupyter notebook implementing an ensemble model for sugarcane disease prediction.
• Combined multiple deep learning architectures: MobileNet, InceptionV3, VGG16, Conv2D, and ResNet152 into a cohesive model.
• Implemented data preprocessing steps, including image augmentation and normalization.
• Conducted model training and evaluation, achieving an overall accuracy of 90.94%.
• Included detailed comments and explanations throughout the notebook for clarity and reproducibility.
• Visualized model performance metrics and comparison graphs for individual models.

Checklist

• I have tested the changes locally
• Documentation has been updated
• Changes are backward-compatible

Model Evaluation

Model	Accuracy
MobileNet	83.20%
VGG16	79.76%
Conv2D	86.06%
ResNet152	81.21%
InceptionV3	86.06%
Ensemble	90.94%

Additional Notes

This model provides a robust solution for timely diagnosis of sugarcane diseases, which is crucial for improving agricultural productivity.

[vercel]

The latest updates on your projects. Learn more about Vercel for Git ↗︎

Name	Status	Preview	Comments	Updated (UTC)
agro-tech-ai	✅ Ready (Inspect)	Visit Preview	💬 Add feedback	Oct 6, 2024 6:49pm

[github-actions]

Thank you for submitting your pull request! 🙌 We'll review it as soon as possible. In the meantime, please ensure that your changes align with our CONTRIBUTING.md. If there are any specific instructions or feedback regarding your PR, we'll provide them here. Thanks again for your contribution! 😊

[IkkiOcean]

Please make it level 3 for GSSOC-ext as I have implemented 5-6 deep learning models and used the best of the them to create an ensemble model which can be implemented easily in the application.

Program : GSSOC-ext and hacktoberfest

[manikumarreddyu]

hi @IkkiOcean can you please inlude .h5 or tflite or .pkl file of the model..it will be easy for us to develop frontend and api..

[github-actions]

🎉 Your pull request has been successfully merged! 🎉 Thank you for your valuable contribution to our project. Your efforts are greatly appreciated. Feel free to reach out if you have any more contributions or if there's anything else we can assist you with. Keep up the fantastic work! 🚀

[manikumarreddyu]

hi @IkkiOcean do it in new PR..

[IkkiOcean]

hi @IkkiOcean can you please inlude .h5 or tflite or .pkl file of the model..it will be easy for us to develop frontend and api..

• The problem is the model's size totals up to almost 800mb. But the max file size limit which can be uploaded on git can not be more than 100mb. I did gave the link to download the model from my Kaggle notebook at ./models/readme.md
• There is another way to do it which is using git-lfs(large file system) to upload larger files but my quota of git-lfs has been extinguished. I can share the file and you can upload it to the repo using git-lfs.

[manikumarreddyu]

Okay 👍

[manikumarreddyu]

Upload it to Google drive and share me the link...I will place it in readme file..

[IkkiOcean]

Upload it to Google drive and share me the link...I will place it in readme file..

• Download Here

I have uploaded the model.zip file at the above url with 6 models required to run the ensemble model. The link will expire after 7 days so be sure to download it and proceed with the files.