Inspiration
Did you know more than 60% of India's paddy production is being affected by various diseases and the major reason for a lack of immediate response is that it's hard to notice? well, we built an ai model for that using TensorFlow and deployed it in Django.
What it does
It classifies the given image of paddy and also it has a scraper that collects news related to farming.
How we built it
We built the model using the conventional CNN model which has 8layers and trained using an Adam optimizer. The dataset is available on Kaggle as link
Challenges we ran into
We are just beginners to the world of deep learning and so we struggled a lot for bringing up the accuracy of our model. It is also the first time we ever built a full-fledged web app.
Accomplishments that we're proud of
We ranked 350 out of 500 submissions in Kaggle well we thought it is a great thing :)
What we learned
We learned that it is hard to host a machine learning model and we respect all open source cloud services such as colab, jupyterlab for giving a GPU space so that even students like us having low end GPU can perform deep learning computations
What's next for Agri Buddy - Paddy Doctor
Well we intend to add more models for corn leaf and honey production and host the app on a server so every farmer can acces it
This project is made by Balaji Anbalagan , Aswin ramanathan , Kumaran of Anna University
Built With
- beautiful-soup
- django
- joblib
- numpy
- python
- react
- tensorflow
Log in or sign up for Devpost to join the conversation.