Tomato_Leaf_Disease_Prediction
A DL project with deployment to predict tomato leaf disease using transfer learning techniques
This projects helps predicting sign language gestures. I tried various transfer learning models for training and chose Densenet121 because of comparatively better performance and smaller size benefitial for smoother deployment.
Project Structure
- Tomato_transfer_learning_densenet121.ipynb file gives the walkthrough over the complete project. Weights for all models trained are stored in the models/model_weights folder. The models folder also contain all the ipynb files giving the walkthrough over training of all models at models/model_training location.
- label.txt file contains all the 10 classes to be predicted with model.
- Predicted_Images folder contains all the predicted images and label_save.txt stores its prediction value along with probability of prediction.
- app.py file gives the walkthrough over the deployment of project in flask. All the required templates are stored in templates folder. The info.ini file contains information shown in after prediction.
- test_images folder contains images that can be used for training.
To run the project, follow below steps
- Ensure that you are in the project home directory
- Create anaconda environment
- Activate environment
-
pip install -r requirement.txt
-
python app.py
- Navigate to URL http://localhost:5000
Please feel free to connect for any suggestions or doubts!!!
Credits
- The credits for dataset used for training goes to https://www.kaggle.com/noulam/tomato
- I have modified https://github.com/Pawandeep-prog/resnet-flask-webapp/tree/main/templates html templates for flask
- The credit for image used in html file for background goes to: