The application uses a fine-tuned TensorFlow model (loaded_model) to make predictions based on user input. When the user submits a sentence through the /predict route, the model predicts the likelihood of it being a real disaster. The predicted label and probability are then displayed in the result.html template.
!I didn't share DisasterTFModel in the repo because the model file is too big.
boraeryazgan/DisasterApp
Folders and files
| Name | Name | Last commit date | ||
|---|---|---|---|---|