sthasus/tomato_leaf_disease_detection

How sthasus/tomato_leaf_disease_detection works

Tomato Leaf Disease Detector is a Django application for account-based tomato-leaf diagnosis. A user signs up or logs in, uploads a JPG/PNG leaf image, and receives a predicted class, confidence, care guidance, and optionally an explanation; logged-in users can revisit saved predictions, while model metrics are public. Separately, the documented training workflow produces the model and metrics consumed by the app. The request and inference paths are shown from excerpts; training-to-artifact links follow README documentation, since the notebook was not sampled.

Main entry points: views.py Upload handler · urls.py Detector routes · inference.py Image inference · train.ipynb Model training

Components 19

Web application

Detector routes detector/urls.py
Landing page detector/views.py
Upload handler detector/views.py
Image validation detector/forms.py
Prediction history detector/views.py
Metrics page detector/views.py
Prediction records detector/models.py
Uploaded images media/uploads

Accounts

Account routes accounts/urls.py
Signup handler accounts/views.py
Signup validation accounts/forms.py

Diagnosis

Image inference detector/inference.py
LIME explanation detector/explain.py
Disease guidance detector/disease_info.py

Model lifecycle

Model training ml_model/train.ipynb
Model weights ml_model/saved_model
Class labels ml_model/class_names.json
Model metrics ml_model/metrics.json

Other components

Visitor

Connections 24

Visitor → Detector routes: requests pages detector/urls.py
Detector routes → Landing page: routes to detector/urls.py
Detector routes → Upload handler: routes to detector/urls.py
Detector routes → Prediction history: routes to detector/urls.py
Detector routes → Metrics page: routes to detector/urls.py
Visitor → Account routes: requests account accounts/urls.py
Account routes → Signup handler: routes to accounts/urls.py
Signup handler → Signup validation: uses form accounts/views.py
Upload handler → Image validation: validates upload detector/views.py
Upload handler → Uploaded images: saves image detector/views.py
Upload handler → Image inference: predicts image detector/views.py
Image inference → Model weights: loads weights README.md
Image inference → Class labels: loads labels README.md
Upload handler → Disease guidance: gets care tips detector/views.py
Upload handler → Prediction records: saves result detector/views.py
Upload handler → LIME explanation: optionally explains detector/views.py
LIME explanation → Image inference: reuses model detector/inference.py
Prediction history → Prediction records: reads history detector/views.py
Prediction history → Disease guidance: gets care tips detector/views.py
Metrics page → Model metrics: reads metrics README.md
Metrics page → Disease guidance: formats labels detector/views.py
Model training → Model weights: exports weights README.md
Model training → Class labels: exports labels README.md
Model training → Model metrics: exports metrics README.md

More Jupyter Notebook diagrams: ksm26/ai-agents-in-langgraph · riazml/math-for-llms · bigcode-project/astraios · bowang-lab/agile · footcricket05/srmeduhub

Drawn by GitDiagram on September 30, 2026 from the repository's file tree, README and source files. Read as Markdown ·