proposalbot
7c5b3f5e25
- Reviewed-by: gtema <artem.goncharov@gmail.com> Co-authored-by: proposalbot <proposalbot@otc-service.com> Co-committed-by: proposalbot <proposalbot@otc-service.com>
3.5 KiB
- original_name
modelarts_23_0098.html
Introduction to Model Templates
Because the configurations of models with the same functions are similar, ModelArts integrates the configurations of such models into a common template. By using this template, you can easily and quickly import models without compiling the config.json configuration file. In simple terms, a template integrates AI engine and model configurations. Each template corresponds to a specific AI engine and inference mode. With the templates, you can quickly import models to ModelArts.
Using a Template
The following uses the template described in as an example. Upload the TensorFlow model package to OBS before using the template. Store the model files in the model directory. When creating a model using this template, you need to select the model directory.
On the Import Model page, set Meta Model Source to Template.
In the Template area, select .
ModelArts also provides three filter criteria: Type, Engine, and Environment, helping you quickly find the desired template. If the three filter criteria cannot meet your requirements, you can enter keywords to search for the target template.
Figure 1 Selecting a template For Model Folder, select the model directory where the model files reside. For details, see
Template Description <modelarts_23_0118>.Note
If a training job is executed for multiple times, different version directories are generated, such as V001 and V002, and the generated models are stored in the model folder in different version directories. When selecting model files, specify the model folder in the corresponding version directory.
If the default input and output mode of the selected template can be overwritten, you can select an input and output mode based on the model function or application scenario. Input and Output Mode is an abstract of the API in config.json. It describes the interface provided by the model for external inference. An input and output mode describes one or more APIs, and corresponds to a template.
For details about the supported input and output modes, see
Input and Output Modes <modelarts_23_0099>.
Supported Templates
TensorFlow-py27 General Template <modelarts_23_0161>TensorFlow-py36 General Template <modelarts_23_0162>MXNet-py27 General Template <modelarts_23_0163>MXNet-py37 General Template <modelarts_23_0164>PyTorch-py27 General Template <modelarts_23_0165>PyTorch-py37 General Template <modelarts_23_0166>Caffe-CPU-py27 General Template <modelarts_23_0167>Caffe-GPU-py27 General Template <modelarts_23_0168>Caffe-CPU-py37 General Template <modelarts_23_0169>Caffe-CPU-py37 General Template <modelarts_23_0169>
Supported Input and Output Modes
Built-in Object Detection Mode <modelarts_23_0100>Built-in Image Processing Mode <modelarts_23_0101>Built-in Predictive Analytics Mode <modelarts_23_0102>Undefined Mode <modelarts_23_0103>