liutension 07fc5a9bb5 | 3 years ago | |
---|---|---|
.. | ||
README.md | 3 years ago | |
app.py | 3 years ago | |
dockerfile | 3 years ago |
Open-Exchange is a model conversion and visualization tool to help users inter-operate among different deep learning frameworks. Convert models between Keras, MXNet and PyTorch.
Configure the rest server ip and port in config/exchange-config.yaml.
1.POST source_framwork to IR (Intermediate Representation)
Submit the model for model conversion
*Request*
```
POST /api/v1/exchange
```
*Parameters*
```
{
* parameters *
"source_framwork": "MXNet/Keras/PyTorch",
"destination_framework": "IR",
"input_shape": "The input shape of the model",
"output_path": "/path/to/model",
* parameters for MXNet-to-IR *
"json_file_path": "The json file for structure of the MXNet model",
"params_file_path": "The params file for weights of the MXNet model",
* parameters for Keras-to-IR *
"json_file_path": "The json file for structure of the Keras model",
"model_file_path": "The file for the Keras model (filename.h5)",
* parameters for Pytorch-to-IR *
"model_file_path": "The file for the Pytorch model" (use torch.save() to generate),
* optional parameters *
"model" : {
"doc_url": "Link to the description document for the model",
"contributor_name": "List of names of the model's contributors",
"contributor_email": "Contact information of the model's contributors",
"contributor_institute": "Institute of the model's contributors",
"framework_name": "Initial training framework",
"framework_version": "The version of initial training framework",
"model_name": "The name of the model",
"model_version": "The version of the model"
}
}
```
*Response if succeeded*
```
Status: 201
{
"response": " "source_framwork" to "doel_framework" success! ",
"json_file_name": "The json file for structure of IR model"
"proto_file_name": "The proto file for structure of IR model"
"weights_file_name": "The weights file for structure of IR model"
}
```
*Response if an error occured*
```
Status: 401
{
"error": "OperatorError",
"message": "The RNN arguments of input are not supported"
}
```
2.POST IR (Intermediate Representation) to destination_framework
Submit the model for model conversion
*Request*
```
POST /api/v1/exchange
```
*Parameters*
```
{
* parameters *
"source_framwork": "IR",
"destination_framework": "MXNet/Keras/PyTorch",
"output_path": "/path/to/model",
* parameters for IR *
"proto_file_path": "The proto file for structure of IR model",
"weights_file_path": "The weights file for structure of IR model"
}
```
*Response if succeeded*
```
Status: 201
{
"response": " "source_framwork" to "doel_framework" success! ",
"weights_file_name": "The weights file for structure of destination_framework model",
}
```
*Response if an error occured*
```
Status: 401
{
"error": "OperatorError",
"message": "The RNN arguments of input are not supported"
}
```
启智章鱼项目(OPENI-OCTOPUS)是一个集群管理和资源调度系统,支持在GPU集群中运行AI任务作业(比如深度学习任务作业)。平台提供了一系列接口,能够支持主流的深度学习框架。
JavaScript Go SVG Python JSX other
Dear OpenI User
Thank you for your continuous support to the Openl Qizhi Community AI Collaboration Platform. In order to protect your usage rights and ensure network security, we updated the Openl Qizhi Community AI Collaboration Platform Usage Agreement in January 2024. The updated agreement specifies that users are prohibited from using intranet penetration tools. After you click "Agree and continue", you can continue to use our services. Thank you for your cooperation and understanding.
For more agreement content, please refer to the《Openl Qizhi Community AI Collaboration Platform Usage Agreement》