From d16d37f203be592b1a0bd8178184845f2e279062 Mon Sep 17 00:00:00 2001 From: Aherobo Ovie Victor Date: Wed, 16 Jul 2025 17:48:27 +0100 Subject: [PATCH] update, added the models and the processed folders --- README.md | 8 ++++---- src/config.py | 2 +- 2 files changed, 5 insertions(+), 5 deletions(-) diff --git a/README.md b/README.md index e0f4b5f..4c04419 100644 --- a/README.md +++ b/README.md @@ -228,10 +228,10 @@ Alternatively, you can use Docker to run the entire system: The Random Forest model achieves the following performance metrics on the validation set: -- **Accuracy**: ~99.5% -- **Precision**: ~95% (minimizing false positives) -- **Recall**: ~92% (minimizing false negatives) -- **F1 Score**: ~93% (balance between precision and recall) +- **Accuracy**: ~99.84% +- **Precision**: ~94.78% (minimizing false positives) +- **Recall**: ~77.35% (minimizing false negatives) +- **F1 Score**: ~85.18% (balance between precision and recall) The most important features for fraud detection include: 1. Transaction amount diff --git a/src/config.py b/src/config.py index e80b781..a5b8be7 100644 --- a/src/config.py +++ b/src/config.py @@ -22,7 +22,7 @@ MODEL_METADATA_PATH = MODELS_DIR / 'model_metadata.json' # API settings API_HOST = '0.0.0.0' -API_PORT = 8000 +API_PORT = 8001 # Web UI settings WEB_HOST = '0.0.0.0'