README.md

---
license: apache-2.0
pipeline_tag: object-detection
tags:
- pytorch
- torch-dag
---
# Model Card for yolov8n_pruned_59

This is a prunned version of the [YOLOv8n](https://github.com/ultralytics/ultralytics#models) model in a [toch-dag](https://github.com/TCLResearchEurope/torch-dag) format.

This model has rougly 59% of the original model FLOPs with small metrics drop.


| Model         | KMAPPs* | M Parameters | mAP50-95 (640x640) |
| -----------   | ------- | ------------ | ------------------ |
| **YOLOv8n (baseline)**   | 21.5 | 3.16 | 37.3 |
| **yolov8n_pruned_59 (ours)**  | 12.6 **(59%)** | 2.00 **(63%)** | 33.3 **(↓ 4)** |


\***KMAPPs** thousands of FLOPs per input pixel

`KMAPPs(model) = FLOPs(model) / (H * W * 1000)`, where `(H, W)` is the input resolution.

The accuracy was calculated on the  COCO val2017 dataset. For details about image pre-processing, please refer to the original repository.
## Model Details

### Model Description


- **Developed by:** [TCL Research Europe](https://github.com/TCLResearchEurope/)
- **Model type:** Object detection
- **License:** Apache 2.0
- **Finetuned from model:** [YOLOv8n](https://github.com/ultralytics/ultralytics#models)

### Model Sources
- **Repository:** [YOLOv8n](https://github.com/ultralytics/ultralytics#models)



## How to Get Started with the Model

To load the model, You have to install [torch-dag](https://github.com/TCLResearchEurope/torch-dag#3-installation) library, which can be done using `pip` by

```
pip install torch-dag
```

then, clone this repository

```
# Make sure you have git-lfs installed (https://git-lfs.com)
git lfs install
git clone https://huggingface.co/TCLResearchEurope/yolov8n_pruned_59
```

and now You are ready to load the model:

```
import torch_dag
import torch

model = torch_dag.io.load_dag_from_path('./yolov8n_pruned_59')

model.eval()
out = model(torch.ones(1, 3, 224, 224))
print(out.shape)
```