---
template: hub1
title: retinanet
summary:
en_US: RetinaNet pre-trained on COCO
zh_CN: RetinaNet (COCO预训练权重)
author: MegEngine Team
tags: [vision, detection]
github-link: https://github.com/megengine/models
---
```python
from megengine import hub
model = hub.load(
"megengine/models",
"retinanet_res50_1x_800size",
pretrained=True,
)
model.eval()
models_api = hub.import_module(
"megengine/models",
git_host="github.com",
)
```
所有预训练模型希望数据被正确预处理。
模型要求输入BGR的图片, 同时需要等比例缩放到:短边和长边分别不超过800/1333
最后做归一化处理 (均值为: `[103.530, 116.280, 123.675]`, 标准差为: `[57.375, 57.120, 58.395]`)。
下面是一段处理一张图片的样例代码。
```python
# Download an example image from the megengine data website
import urllib
url, filename = ("https://data.megengine.org.cn/images/cat.jpg", "cat.jpg")
try: urllib.URLopener().retrieve(url, filename)
except: urllib.request.urlretrieve(url, filename)
# Read and pre-process the image
import cv2
image = cv2.imread("cat.jpg")
data, im_info = models_api.DetEvaluator.process_inputs(image, 800, 1333)
model.inputs["image"].set_value(data)
model.inputs["im_info"].set_value(im_info)
from megengine import jit
@jit.trace(symbolic=True)
def infer():
predictions = model(model.inputs)
return predictions
print(infer())
```
### 模型描述
目前我们提供了retinanet的预训练模型, 在coco验证集上的结果如下:
| model | mAP
@5-95 |
| --- | --- |
| retinanet-res50-1x-800size | 36.0 |
### 参考文献
- [Focal Loss for Dense Object Detection](https://arxiv.org/pdf/1708.02002) Tsung-Yi Lin, Priya Goyal, Ross Girshick, Kaiming He, Piotr Dollár. Proceedings of the IEEE international conference on computer vision. 2017: 2980-2988.
- [Microsoft COCO: Common Objects in Context](https://arxiv.org/pdf/1405.0312.pdf) Lin, Tsung-Yi and Maire, Michael and Belongie, Serge and Hays, James and Perona, Pietro and Ramanan, Deva and Dollár, Piotr and Zitnick, C Lawrence
Lin T Y, Maire M, Belongie S, et al. European conference on computer vision. Springer, Cham, 2014: 740-755.
All pre-trained models expect input images normalized in the same way,
i.e. input images must be 3-channel BGR images of shape `(H x W x 3)`, and reszied shortedge/longedge to no more than `800/1333`.
The images should be normalized using `mean = [103.530, 116.280, 123.675]` and `std = [57.375, 57.120, 58.395])`.
Here's a sample execution.
```python
# Download an example image from the megengine data website
import urllib
url, filename = ("https://data.megengine.org.cn/images/cat.jpg", "cat.jpg")
try: urllib.URLopener().retrieve(url, filename)
except: urllib.request.urlretrieve(url, filename)
# Read and pre-process the image
import cv2
image = cv2.imread("cat.jpg")
data, im_info = models_api.DetEvaluator.process_inputs(image, 800, 1333)
model.inputs["image"].set_value(data)
model.inputs["im_info"].set_value(im_info)
from megengine import jit
@jit.trace(symbolic=True)
def infer():
predictions = model(model.inputs)
return predictions
print(infer())
```
### Model Description
Currently we provide a `retinanet` model which is pretrained on `COCO2017` training set. The mAP on `COCO2017` val set can be found in following table.
| model | mAP
@5-95 |
| --- | --- |
| retinanet-res50-1x-800size | 36.0 |
### References
- [Focal Loss for Dense Object Detection](https://arxiv.org/pdf/1708.02002) Tsung-Yi Lin, Priya Goyal, Ross Girshick, Kaiming He, Piotr Dollár. Proceedings of the IEEE international conference on computer vision. 2017: 2980-2988.
- [Microsoft COCO: Common Objects in Context](https://arxiv.org/pdf/1405.0312.pdf) Lin, Tsung-Yi and Maire, Michael and Belongie, Serge and Hays, James and Perona, Pietro and Ramanan, Deva and Dollár, Piotr and Zitnick, C Lawrence
Lin T Y, Maire M, Belongie S, et al. European conference on computer vision. Springer, Cham, 2014: 740-755.