Skip to content
GitLab
Menu
Projects
Groups
Snippets
Loading...
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in / Register
Toggle navigation
Menu
Open sidebar
ModelZoo
ResNet50-v2_tvm
Commits
6860a6b1
Commit
6860a6b1
authored
Jun 16, 2023
by
zhanggzh
Browse files
readme file
parent
6a11481d
Changes
1
Show whitespace changes
Inline
Side-by-side
Showing
1 changed file
with
56 additions
and
28 deletions
+56
-28
README.md
README.md
+56
-28
No files found.
README.md
View file @
6860a6b1
# TVM
# TVM
## 模型介绍
## 模型介绍
使用深度学习编译器TVM对ResNet50网络模型进行推理及调优
```
使用深度学习编译器TVM对ResNet50网络模型进行推理、调优及部署
```
## 模型结构
## 模型结构
ResNet50-v2
```
ResNet50-v2
```
## 数据集及模型文件
## 数据集及模型文件
模型文件下载地址:
模型文件下载地址:
```
"https://github.com/onnx/models/raw/main/vision/classification/resnet/model/resnet50-v2-7.onnx"
"https://github.com/onnx/models/raw/main/vision/classification/resnet/model/resnet50-v2-7.onnx"
## 推理及自动调优
```
## 推理、自动调优及部署
### 环境配置
### 环境配置
拉取镜像:
拉取镜像:
```
docker pull image.sourcefind.cn:5000/dcu/admin/base/custom:tvm-0.10_dtk-22.10_py38_centos-7.6
```
### 执行推理及调优
### 执行推理及调优
下载模型文件后执行以下命令进行推理测试及调优测试:
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
python tune_resnet50-v2.py
```
### 部署推理
下载配置好镜像之后,将/tvm-0.11-dev0/apps/howto_deploy、路径下原有代码移除,将prepare_test_libs.py、run_example.sh、tvm_runtime_pack.cc、cow.jpg、Makefile、resnet50-v2_deploy.cc等代码及图片放置到该路径下执行
以下命令:
```
mkdir -p lib
python prepare_test_libs.py
sh run_example.sh
```
## 准确率数据
```
max_num:15.6692
max_iter:0x28cda14
max_num_index:345
```
## TVM版本
## TVM版本
TVM-0.10
```
TVM-0.11
```
## 源码仓库及问题反馈
* https://developer.hpccube.com/codes/modelzoo/tvm_tune_resnet50-v2
## 历史版本
## 历史版本
* https://developer.hpccube.com/codes/modelzoo/tvm_tune_resnet50-v2
* https://developer.hpccube.com/codes/modelzoo/tvm_tune_resnet50-v2
## 参考
## 参考
...
...
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
.
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment