Skip to content

Commit c08b001

Browse files
Update README.md
1 parent 67305c0 commit c08b001

File tree

1 file changed

+49
-2
lines changed

1 file changed

+49
-2
lines changed

README.md

Lines changed: 49 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,2 +1,49 @@
1-
# Find-Card
2-
Find-Card
1+
# DetectX: Simple Yet Powerful Object Detection Framework
2+
3+
A lightweight and modular object detection framework powered by Detectron2, focusing on easy training and deployment.
4+
5+
## Core Features
6+
7+
- 🎯 Pre-configured Detectron2 models (Faster R-CNN, RetinaNet)
8+
- 🔄 Simple data pipeline for custom datasets
9+
- 📊 Built-in evaluation metrics (COCO metrics, RMSE, MSE, PSNR)
10+
- 🚀 Easy model configuration and training
11+
12+
## Quick Start
13+
14+
```bash
15+
# Install dependencies
16+
pip install -r requirements.txt
17+
18+
# Train a model
19+
from Model.modelling.detectron2 import Detectron2
20+
21+
model = Detectron2(
22+
model="COCO-Detection/faster_rcnn_X_101_32x8d_FPN_3x.yaml",
23+
max_iter=500,
24+
base_lr=0.00025
25+
)
26+
model.train()
27+
```
28+
29+
## Project Structure
30+
31+
```
32+
ML/
33+
├── Model/
34+
│ ├── modelling/ # Core model implementations
35+
│ ├── dataset/ # Dataset handling utilities
36+
│ └── metrics/ # Evaluation metrics
37+
└── tests/ # Unit tests
38+
```
39+
40+
## Currently Supported
41+
42+
- Models: Faster R-CNN, RetinaNet
43+
- Metrics: COCO AP, RMSE, MSE, PSNR
44+
- Data formats: COCO-style annotations
45+
- GPU acceleration with CUDA
46+
47+
## License
48+
49+
Apache License 2.0

0 commit comments

Comments
 (0)