nikos99n commited on
Commit
d29ed02
Β·
verified Β·
1 Parent(s): fe66586

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +8 -29
README.md CHANGED
@@ -1,29 +1,8 @@
1
- # By Kolia Aimilia, Kontoudakis Nikos, Skiada Kyriaki, Lampropoulou Nancy
2
- The project aims to classify lesions.
3
-
4
- # Project Structure
5
- The project follows a modular structure to separate configuration, data processing, computer vision logic, and model training.
6
-
7
- ```text
8
- ham10000_project/
9
- β”‚
10
- β”œβ”€β”€ data/
11
- β”‚ β”œβ”€β”€ images/ # Original dermoscopy images (e.g., ISIC_0024306.jpg)
12
- β”‚ └── GroundTruth.csv # Metadata and labels
13
- β”‚
14
- β”œβ”€β”€ models/ # Generated automatically during training
15
- β”‚ β”œβ”€β”€ skin_cancer_model.pkl # Trained Random Forest/SVM model
16
- β”‚ β”œβ”€β”€ scaler.pkl # StandardScaler for feature normalization
17
- β”‚ β”œβ”€β”€ classes.pkl # List of class names (MEL, NV, etc.)
18
- β”‚ └── comparison_results.png # Confusion matrix plot
19
- β”‚
20
- β”œβ”€β”€ src/ # Core Logic Package
21
- β”‚ β”œβ”€β”€ __init__.py # Makes this folder a Python package
22
- β”‚ β”œβ”€β”€ config.py # Configuration, Constants, and Hyperparameters
23
- β”‚ β”œβ”€β”€ data.py # Data loading and stratified splitting logic
24
- β”‚ β”œβ”€β”€ features.py # Computer Vision pipeline (CLAHE, Otsu, Sobel, etc.)
25
- β”‚ └── model.py # Model training, evaluation, and saving
26
- β”‚
27
- β”œβ”€β”€ train_main.py # Script 1: Main entry point to train the model
28
- β”œβ”€β”€ app.py # Script 2: Streamlit Web Interface for inference
29
- └── requirements.txt # Project dependencies
 
1
+ title: Lesion Dection
2
+ emoji: 🐒
3
+ colorFrom: blue
4
+ colorTo: green
5
+ sdk: streamlit
6
+ sdk_version: 1.45.1
7
+ app_file: ./app.py
8
+ pinned: false