---
base_model:
- ultralytics/yolo11
model_name: yolo11-blood_cell-onnx
tags:
- object-detection
- sft
- yolov11
- medical-imaging
- blood-cell
- fp16
license: apache-2.0
language:
- en
pipeline_tag: object-detection
library_name: ultralytics
---

# 🩸 YOLO11 — Blood Cell Detector (Freeze-10, FP16 ONNX)

This model is a **fine-tuned YOLO11** trained for **blood cell detection** on the  
[Blood Cell Dataset (Roboflow)](https://universe.roboflow.com/med-2fnb4/blood-cell-dataset-qbrgd).  
The first **10 layers were frozen** to retain pretrained spatial features, and the model was exported to **ONNX (FP16)** for efficient inference.

---

## ⚙️ Configuration

| Attribute | Value |
|------------|-------|
| **Base Model** | `yolo11n.pt` |
| **Dataset** | [Blood Cell (Roboflow)](https://universe.roboflow.com/med-2fnb4/blood-cell-dataset-qbrgd) |
| **Epochs** | 30 |
| **Batch Size** | 32 |
| **Image Size** | 640×640 |
| **Optimizer** | Auto |
| **Freeze Layers** | 10 |
| **Precision** | FP16 (half=True) |
| **Export Format** | ONNX |
| **Device** | GPU (0,1) |

---

## 🩺 Example Detection

![Blood Cell Detection](https://huggingface.co/Jesteban247/yolo11-blood_cell-onnx/resolve/main/Blood.png)

---

## 📈 Results

| Metric | Value |
|---------|-------|
| **mAP50** | 0.974 |
| **mAP50-95** | 0.905 |
| **Precision (B)** | 0.951 |
| **Recall (B)** | 0.920 |
| **Inference Time (ms)** | 33.73 |
| **FPS** | 29.65 |
| **Model Size (MB)** | 5.2 |

> FP16 inference maintained **identical accuracy** to FP32 while reducing latency.  
> Layer freezing improved training stability and avoided overfitting on the limited dataset.