Commit e79f2ee3 authored by dcuai's avatar dcuai
Browse files

Update README.md

parent d55c8bd2
......@@ -10,11 +10,7 @@ ResNet50网络中包含了49个卷积层、1个全连接层等
ResNet50使用了多个具有残差连接的残差块来解决梯度消失或梯度爆炸问题,并使得网络可以向更深层发展。
![image-1.png](./image-1.png)
## 模型文件
模型文件下载地址:
```
"https://github.com/onnx/models/raw/main/vision/classification/resnet/model/resnet50-v2-7.onnx"
```
## 环境配置
### Docker(方法一)
......@@ -41,8 +37,15 @@ source /opt/dtk/env.sh
C++部署代码使用数据为:
```
"https://github.com/ultralytics/yolov5/releases/download/v1.0/coco128.zip"
coco128数据集的目录结构如下:
```
## 推理
模型文件下载地址:
```
"https://github.com/onnx/models/raw/main/vision/classification/resnet/model/resnet50-v2-7.onnx"
```
### 推理
下载模型文件后执行以下命令进行推理测试及调优测试:
```
python tune_resnet50-v2.py
......@@ -57,7 +60,10 @@ source /opt/dtk/env.sh
python prepare_test_libs.py
sh run_example.sh
```
## 精度
## result
### 精度
```
max_num:15.6692
max_iter:0x28cda14
......@@ -77,7 +83,7 @@ source /opt/dtk/env.sh
* https://developer.hpccube.com/codes/modelzoo/tvm_tune_resnet50-v2
## 参考
## 参考资料
* [https://tvm.apache.org/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html]()
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment